The Man Who Calls BS On AI: They’re LYING About AI, 2027 Is When It All Breaks! | Ed Zitron

TL;DR

• Ed Zitron (tech PR veteran, "Better Offline" podcast, ~16 years in the industry) tells Steven Bartlett that generative AI is "at its heart a con" — not that the software does nothing, but that its capabilities and financials are systematically overstated: "the largest non-consensual push of technology in history."
• The financial core of his case: every AI company runs at a horrifying loss (OpenAI lost $20.9B last year); ~70% of AI revenue flows from two unprofitable companies (OpenAI, Anthropic) that are funded by the very hyperscalers reporting the revenue; a $200/month subscription can burn $8-14k of real token cost; a trillion-plus of capex chases ~$22B of non-OpenAI/Anthropic revenue.
• His "rot-com bubble" theory: big tech ran out of hypergrowth ideas, bought GPUs, and the market rewarded the spending as if it were proof of demand — circular financing (Nvidia→CoreWeave, hyperscalers→OpenAI/Anthropic) does the rest. Zuckerberg's "invest aggressively" = "Shrek's Lord Farquaad: some of you may die, but that's a risk I'm willing to accept."
• The prediction: OpenAI runs out of cash around 2027 (can't IPO before Anthropic, needs $100B+/year to survive), triggering restated hyperscaler guidance, Nvidia revenue down 50-70%, a "tech depression," and 20-40% off retirement-heavy indexes. His hedge: he'd be wrong given a ~1000x hardware cost breakthrough plus genuinely autonomous capability.
• Bartlett pushes back hard throughout — adoption stats, hallucination-rate improvements, the Ballmer-iPhone-laughing clip, innovator's-dilemma analogies, autonomous-vehicle safety data — making this one of the more genuinely adversarial DOAC episodes; Zitron's consistent reply: "if it was going well, they'd tell you," and demand-at-honest-prices is the only test that matters.

Key moments

• [02:38] The thesis — generative AI is "at its heart a con": overstated capabilities, overstated financials, exploiting weaknesses in journalism and markets (frame_0001.jpg — DOAC studio, AI-company boxes as props on the table, frame_0017.jpg).
• [08:02] Adoption isn't consent — Gemini in Docs, Copilot in Word, Rufus on Amazon: "the largest non-consensual push of technology in history."
• [12:00] Tokens explained — the taxi-meter economics; SemiAnalysis found $200/month ChatGPT subs burning $14k of tokens; Uber blew its annual token budget in 3 months when honest enterprise pricing arrived ([13:53]).
• [24:23] The hallucination fight — Bartlett cites a leaderboard (21.8%→0.7% on simple summarization); Zitron counters with his Bloomberg-terminal fabricated stock price and "the more detailed the report, the more likely something's wrong."
• [31:31] Matt Hughes vs the machine — why he pays a human editor: context, trust, shared learning; Bartlett's rebuttal: people pay for output, not process.
• [36:10] The iPhone test — real disruptive tech is obvious (Ballmer laughing clip [36:48]); AI still needs "a Pee-wee's Playhouse of harnesses and prompts" to justify itself.
• [42:35] Rot-com bubble — dot-com had dark fiber with post-bubble utility; GPUs have no post-bubble story; today's demand is subsidized, so the analogy fails.
• [52:46] The Google enshittification story — Prabhakar Raghavan, the 2019 "code yellow," worse search to drive more queries, and AI Overviews as "the ultimate form of Google's evil."
• [77:28] The myth-card game — one-sentence rebuttals (frame_0025.jpg, frame_0033.jpg): no economic growth in the data; "what AI race?"; job replacement "just isn't happening" — the disrupted are art directors, transcribers, translators with bosses who never cared about output quality ([66:40]).
• [100:38] Blue-shirt Ed — Zuckerberg's 15-basis-points retention gain from AI content analysis: "10-something billion dollars in and the best you've got is 0.15%?"
• [120:50] Blackmail myths debunked — the TaskRabbit and Anthropic blackmail stories were prompted/trained scenarios covered as autonomy; "mysticism attempts" that backfired ([123:52]).
• [128:27] The 2027 crash sequence — OpenAI can't raise → can't IPO after Anthropic → SoftBank's $100B paper position freezes → restated guidance → Nvidia -50-70% → tech depression ([133:41]); Oracle "probably dies" ([115:11]).
• [139:05] Advice to Jenny and Dave — he lives in cash, calls the market "a casino pumped up by the media"; softened to: be suspicious of tech promises, take gains ([139:39]).
• [145:02] The closing question — surprisingly tender: community among the critics, "reach out to someone you love and tell them their shit rocks."

Hook microscope (0-10s)

• Frames: 20 at 2 fps
• Word-level transcript (32 words):

  [  0.00s] I
  [  0.20s] think
  [  0.50s] generative
  [  1.02s] AI
  [  1.52s] is
  [  2.08s] at
  [  2.22s] its
  [  2.56s] heart
  [  2.78s] con
  [  3.10s] and
  [  3.28s] seeing
  [  3.82s] these
  [  4.20s] ultra
  [  4.66s] rich
  [  5.06s] ultra
  [  5.26s] powerful
  [  5.80s] people
  [  6.04s] lie
  [  6.40s] through
  [  6.54s] their
  [  6.60s] f***ing
  [  7.04s] teeth
  [  7.18s] turns
  [  7.54s] my
  [  7.78s] stomach
  [  8.60s] The
  [  8.62s] word
  [  8.82s] con
  [  9.10s] is
  [  9.24s] a
  [  9.46s] strong
  [  9.64s] word

Hook pattern: DOAC contrarian supercut. Ninety seconds of Zitron's spiciest lines stitched before the intro — "generative AI is at its heart a con," "they are misleading the entire world," the Shrek/Lord Farquaad jab at Zuckerberg — with Bartlett's on-camera astonishment ("you are the first person I've spoken to that has that opinion") positioning the episode as heresy against every prior DOAC AI guest. The myth-card game is teased up front ("we're going to play a game, Ed") as a structural promise. Visuals: dramatic studio lighting, giant keyword overlays, the AI-company boxes lined up on the table like defendants (frame_0017.jpg). Followed immediately by a 55-second subscribe appeal — the DOAC formula at full strength.

Editorial profile

• Shots: 80
• Cuts/min: 0.54
• Mean shot length: 110.87s
• Median shot length: 2.12s
• Talking-head ratio: n/a (opencv not installed)

Premium studio debate-podcast (DOAC formula): moody two-shot with dramatic lighting, giant animated keyword overlays ("RUN AT A LOSS", "GOD"), physical props (AI-company boxes, myth cards), archival clip inserts (Ballmer laughing), a cold-open supercut, and two mid-roll host-read ads plus two subscribe appeals.

Quotable moments

• [00:32] "This is the largest non-consensual push of technology in history."
• [03:19] "What do you call something where from the very beginning they've sold it in the terms of magic, but it's just a half-assed sorcery machine?"
• [72:05] "Every single scam and con starts with rushing you... if anyone tries to rush you and it's not literally a mortal thing, slow down."
• [75:28] "What if these models fall into the wrong hands? They're already in the wrong hands. Mark Zuckerberg, Sam Altman, Dario Amodei."
• [103:24] "If it was going well, you'd tell me how well it was going — rather than doing this weird rain dance where if we move all the pieces around, in 3 years, theoretically, this will happen."
• [127:01] "AI boosters can't speak in the future tense anymore. You get two weeks in the future, max — because constrained to what's happening today, they would sound like insane people."

Entities mentioned

• People: ed-zitron, steven-bartlett (host), sam-altman ("Clammy Sammy"), dario-amodei, mark-zuckerberg, satya-nadella, sundar-pichai, jensen-huang, prabhakar-raghavan, gary-marcus, geoffrey-hinton, ed-elson (Prof G Markets), matt-hughes (his editor), carl-brown (Internet of Bugs), steve-ballmer (archival clip)
• Companies: openai, anthropic, nvidia, microsoft, google, amazon, meta, oracle, softbank, coreweave, broadcom, uber, waymo, zoox, semianalysis, bloomberg
• Tools / products: chatgpt, claude, gemini, copilot, fable (enterprise per-token pricing complaint), stargate-abilene, bloomberg-terminal (AskB/BQL), better-offline (his podcast), wheres-your-ed-at (his Ghost newsletter)
• Places: abilene-texas (Stargate), bristol (power comparison), san-francisco, las-vegas

Concepts surfaced

• rot-com-bubble: Zitron's thesis — big tech ran out of hypergrowth ideas, so GPU capex became a narrative product; the market rewards the spending itself as proof of demand.
• circular-financing: Nvidia invests in CoreWeave which rents Nvidia GPUs back; hyperscalers fund OpenAI/Anthropic whose spend becomes the hyperscalers' "AI revenue" — demand manufactured inside the loop.
• subsidized-demand-test: adoption means nothing until users pay honest per-token costs; the moment enterprises did (March 2026), they revolted — the only demand signal that counts.
• non-consensual-adoption: AI usage stats inflated by forced integration (Gemini in Docs, Copilot in Word) plus 3 years of "use it or be left behind" media pressure.
• obviousness-test: real disruptive tech (iPhone, AWS) needs no explanation — you demo it and people get it; needing a "Pee-wee's Playhouse of harnesses" is disqualifying.
• annualized-run-rate: deliberately undefined revenue metric (month×12? 4 weeks×13?) built to let outlets print big numbers — disclosure-avoidance as marketing.
• cult-of-the-wealthy: "they wouldn't spend a trillion for no reason" — the psychological backstop of every bubble; reconciling that elites are lucky rather than brilliant is grim, so people assume they're missing something.
• future-tense-ban: his proposed booster regulation — describe only what the tech does today (two weeks ahead max) and the pitch collapses.
• mysticism-marketing: blackmail/escape stories are prompted scenarios reported as autonomy, meant to make LLMs seem unknowably powerful — fear as a sales channel that has now backfired.
• business-idiot-theory: bosses who don't do the work love the "ingratiation machine"; workers AI-wash their output to survive performance politics.
• easy-things-easy-hard-things-harder: (via Carl Brown) LLMs excel at small distinct tasks and make complex ones riskier — the multiplicative error problem.

Transcript

Source: captions.

[00:03] I think generative AI is at its heart con and seeing these ultra rich ultra
[00:05] con and seeing these ultra rich ultra powerful people lie through their teeth
[00:07] powerful people lie through their teeth turns my stomach. The word con is a
[00:10] turns my stomach. The word con is a strong word.
[00:10] strong word. >> Well, what do you call something where
[00:12] >> Well, what do you call something where from the very beginning they've sold
[00:13] from the very beginning they've sold [music] it in the terms of magic but
[00:15] [music] it in the terms of magic but it's just a halfass arcery machine. They
[00:17] it's just a halfass arcery machine. They are misleading the entire world.
[00:18] are misleading the entire world. >> You are the first person that I've
[00:19] >> You are the first person that I've spoken to that has that opinion.
[00:21] spoken to that has that opinion. >> Well, the fact that this is happening is
[00:23] >> Well, the fact that this is happening is insane and the fact it's not a scandal
[00:25] insane and the fact it's not a scandal is insane. And I've been in the tech
[00:26] is insane. And I've been in the tech industry for 16 years now and I love
[00:28] industry for 16 years now and I love technology and I'm enthusiastic about
[00:30] technology and I'm enthusiastic about it, but I don't like being misled. And
[00:33] it, but I don't like being misled. And this is the largest non-consensual push
[00:35] this is the largest non-consensual push of technology in history.
[00:37] of technology in history. >> So, we're going to play a game, Ed. I
[00:38] >> So, we're going to play a game, Ed. I have the things that you consider to be
[00:40] have the things that you consider to be myths about the AI industry.
[00:42] myths about the AI industry. >> Let's play it. The AI industry is
[00:44] >> Let's play it. The AI industry is creating enormous economic growth. No,
[00:46] creating enormous economic growth. No, it's not. All of these companies run at
[00:48] it's not. All of these companies run at a horrifying loss. Open AI lost $20.9
[00:51] a horrifying loss. Open AI lost $20.9 billion last year. None of these people
[00:53] billion last year. None of these people can just say, "Yeah, we're on the path
[00:54] can just say, "Yeah, we're on the path to making this profitable." because they
[00:55] to making this profitable." because they can't.
[00:56] can't. >> Next one.
[00:56] >> Next one. >> AI will replace all human jobs. That
[00:59] >> AI will replace all human jobs. That just isn't happening and there's no
[01:00] just isn't happening and there's no economic data to support it. Next, the
[01:02] economic data to support it. Next, the United States need to spend trillions to
[01:04] United States need to spend trillions to beat China in the AI race. What's the
[01:06] beat China in the AI race. What's the race to do for us to constantly piss our
[01:08] race to do for us to constantly piss our pants worrying about China? But people
[01:10] pants worrying about China? But people keep saying, "What if these models fall
[01:11] keep saying, "What if these models fall into the wrong hands? They're already in
[01:13] into the wrong hands? They're already in the wrong hands." Mark Zuckerberg, Sam
[01:15] the wrong hands." Mark Zuckerberg, Sam Olman, Dario Amade.
[01:16] Olman, Dario Amade. >> Mark Zuckerberg says, "We'll continue to
[01:18] >> Mark Zuckerberg says, "We'll continue to invest aggressively in infrastructure to
[01:20] invest aggressively in infrastructure to meet the demand." God met as a
[01:22] meet the demand." God met as a monstrosity. Makes me think of Shrek
[01:24] monstrosity. Makes me think of Shrek with L fogquad. Some of you may die, but
[01:26] with L fogquad. Some of you may die, but that's a risk I'm willing to accept. If
[01:28] that's a risk I'm willing to accept. If only these people gave a about
[01:29] only these people gave a about poverty or actual problems in the world
[01:31] poverty or actual problems in the world versus are we buying enough GPUs. If
[01:34] versus are we buying enough GPUs. If this continues, [music] what does the
[01:36] this continues, [music] what does the future look like?
[01:43] This is super interesting to me. My team given me this report to show me how many
[01:45] given me this report to show me how many of you that watch this show subscribe.
[01:46] of you that watch this show subscribe. And some of you have told us according
[01:48] And some of you have told us according to this that you are unsubscribed from
[01:50] to this that you are unsubscribed from the channel randomly. So, favor to ask
[01:52] the channel randomly. So, favor to ask all of you. Please could you check right
[01:54] all of you. Please could you check right now if you've hit the subscribe button
[01:55] now if you've hit the subscribe button if you are a regular viewer of the show
[01:56] if you are a regular viewer of the show and you like what we do here. We're
[01:58] and you like what we do here. We're approaching quite a significant landmark
[01:59] approaching quite a significant landmark on this show in terms of a subscriber
[02:01] on this show in terms of a subscriber number. So, if there was one simple free
[02:04] number. So, if there was one simple free thing that you could do to help us, my
[02:05] thing that you could do to help us, my team, everyone here to keep this show
[02:07] team, everyone here to keep this show free, to keep it improving year over
[02:09] free, to keep it improving year over year and week over week, it is just to
[02:11] year and week over week, it is just to hit that subscribe button and to double
[02:12] hit that subscribe button and to double check if you've hit it. Only thing I'll
[02:14] check if you've hit it. Only thing I'll ever ask of you, do we have a deal? If
[02:16] ever ask of you, do we have a deal? If you do it, I'll tell you what I'll do.
[02:18] you do it, I'll tell you what I'll do. I'll make sure every single week, every
[02:20] I'll make sure every single week, every single month, we fight harder and harder
[02:21] single month, we fight harder and harder and harder and harder to bring you the
[02:22] and harder and harder to bring you the guests and conversations that you want
[02:24] guests and conversations that you want to hear. I've stayed true to that
[02:25] to hear. I've stayed true to that promise since the very beginning of the
[02:26] promise since the very beginning of the Dire of Sio and I will not let you down.
[02:30] Dire of Sio and I will not let you down. Please help us. Really appreciate it.
[02:31] Please help us. Really appreciate it. Let's get on with the show.
[02:33] Let's get on with the show. [music]
[02:38] >> Ed Zitron, there are a number of things that you
[02:40] there are a number of things that you believe that a lot of other people don't
[02:42] believe that a lot of other people don't believe, right? You have, I think, a
[02:44] believe, right? You have, I think, a couple of controversial opinions and
[02:46] couple of controversial opinions and opinions that are in contrast to the
[02:49] opinions that are in contrast to the other guests that I've sat here with.
[02:51] other guests that I've sat here with. What exactly are those opinions, Ed? I
[02:54] What exactly are those opinions, Ed? I think generative AI is at its heart con.
[02:57] think generative AI is at its heart con. I don't think it is sold as honest
[03:00] I don't think it is sold as honest software. I think that they overstate
[03:02] software. I think that they overstate both what it can do, what it will do,
[03:04] both what it can do, what it will do, and the underlying financials to the
[03:06] and the underlying financials to the point that they are misleading the
[03:08] point that they are misleading the entire world. And they're actively
[03:09] entire world. And they're actively exploiting the weaknesses in journalism,
[03:11] exploiting the weaknesses in journalism, in our economies, and indeed within the
[03:14] in our economies, and indeed within the responsible parties with sellside
[03:16] responsible parties with sellside analysts, governments, and all over the
[03:17] analysts, governments, and all over the shop.
[03:18] shop. >> The word con is a strong word.
[03:20] >> The word con is a strong word. >> Yeah. I mean, what do you call something
[03:22] >> Yeah. I mean, what do you call something where from the very beginning they've
[03:24] where from the very beginning they've sold it in the terms of magic as this
[03:26] sold it in the terms of magic as this thing that will replace all jobs, that
[03:28] thing that will replace all jobs, that will cure cancer, and all of these
[03:29] will cure cancer, and all of these things? And when you look at it, it's
[03:31] things? And when you look at it, it's boring cloud software that's extremely
[03:33] boring cloud software that's extremely expensive and unprofitable and also
[03:35] expensive and unprofitable and also unreliable at its core.
[03:37] unreliable at its core. >> People will be asking where are you
[03:39] >> People will be asking where are you drawing from in terms of your
[03:40] drawing from in terms of your references, your personal experiences?
[03:42] references, your personal experiences? Where were you educate? What you study?
[03:43] Where were you educate? What you study? What you write about? What do you do Ed?
[03:45] What you write about? What do you do Ed? >> So that's the funny thing is people say
[03:46] >> So that's the funny thing is people say he's not got a finance experience. He's
[03:48] he's not got a finance experience. He's not going to take. I've been in the tech
[03:50] not going to take. I've been in the tech industry 15 16 years now in PR but still
[03:53] industry 15 16 years now in PR but still had practical experience and I love
[03:55] had practical experience and I love technology and I'm enthusiastic about
[03:57] technology and I'm enthusiastic about it. And this thing just comes along that
[03:59] it. And this thing just comes along that everyone is telling me is the best thing
[04:00] everyone is telling me is the best thing since sliced bread. And it can't even do
[04:02] since sliced bread. And it can't even do the basics. It can't even do search.
[04:04] the basics. It can't even do search. Well, whenever you ask an AI person,
[04:05] Well, whenever you ask an AI person, well, what's your setup? They describe
[04:07] well, what's your setup? They describe this PeeWee's Playhouse thing of like,
[04:09] this PeeWee's Playhouse thing of like, well, you got to harness here and you
[04:10] well, you got to harness here and you got to use the right prompt. Well, you
[04:12] got to use the right prompt. Well, you don't want to use that prompt. You want
[04:13] don't want to use that prompt. You want to use this prompt here with this model,
[04:15] to use this prompt here with this model, but don't use this model for the
[04:16] but don't use this model for the beginning, but at the end, you're going
[04:18] beginning, but at the end, you're going to want to use this model. And this is
[04:20] to want to use this model. And this is meant to be artificial intelligence.
[04:22] meant to be artificial intelligence. It's meant to be smart. It's meant to be
[04:24] It's meant to be smart. It's meant to be autonomous. It's meant to be something
[04:26] autonomous. It's meant to be something that you set and forget.
[04:27] that you set and forget. >> We have the sort of six leading AI
[04:29] >> We have the sort of six leading AI companies on the table here. Anthropic
[04:31] companies on the table here. Anthropic Amazon, Nvidia, Microsoft, OpenAI,
[04:33] Amazon, Nvidia, Microsoft, OpenAI, Google. You're saying that their
[04:35] Google. You're saying that their fundamental business model is a con.
[04:38] fundamental business model is a con. >> Well, their revenues are not really
[04:39] >> Well, their revenues are not really coming from AI. Up until fairly
[04:41] coming from AI. Up until fairly recently, none of their revenues were
[04:43] recently, none of their revenues were coming from AI. Like dribbles a bit.
[04:45] coming from AI. Like dribbles a bit. Right now, 70% of all AI revenues across
[04:48] Right now, 70% of all AI revenues across those three companies are from OpenAI
[04:50] those three companies are from OpenAI and Anthropic to unprofitable,
[04:52] and Anthropic to unprofitable, unsustainable companies that literally
[04:53] unsustainable companies that literally cannot afford to exist without these
[04:55] cannot afford to exist without these very same companies giving them money.
[04:58] very same companies giving them money. Amazon sent $50 billion to OpenAI this
[05:01] Amazon sent $50 billion to OpenAI this year. They sent $5 billion to Anthropic.
[05:04] year. They sent $5 billion to Anthropic. Google sent $10 billion to Anthropic.
[05:06] Google sent $10 billion to Anthropic. And in the next three and a half years,
[05:08] And in the next three and a half years, OpenAI and Anthropic based on actual
[05:11] OpenAI and Anthropic based on actual sellside analyst evaluations, their
[05:13] sellside analyst evaluations, their estimates that inform whether stock is
[05:15] estimates that inform whether stock is going to go up or down after earnings,
[05:16] going to go up or down after earnings, they are expecting 400 or more billion
[05:19] they are expecting 400 or more billion dollar of revenue, 30 or something% of
[05:23] dollar of revenue, 30 or something% of cloud growth just from these two
[05:25] cloud growth just from these two unprofitable companies that will need to
[05:26] unprofitable companies that will need to be given the money from somewhere. And
[05:28] be given the money from somewhere. And on top of that, these companies have
[05:31] on top of that, these companies have such low respect for the average
[05:33] such low respect for the average investor, for the analyst, for everyone
[05:35] investor, for the analyst, for everyone really that they don't even disclose
[05:37] really that they don't even disclose their AI revenues. The few times they
[05:39] their AI revenues. The few times they dain us worthy, they use something
[05:41] dain us worthy, they use something called a run rate, an annualized run
[05:43] called a run rate, an annualized run rate, which means well, nothing. They
[05:45] rate, which means well, nothing. They never define it. It can mean months 12.
[05:47] never define it. It can mean months 12. It can mean month 13. It can mean last 4
[05:50] It can mean month 13. It can mean last 4 weeks time 13. It's different every
[05:52] weeks time 13. It's different every time, and they never define it. And then
[05:53] time, and they never define it. And then they sometimes just don't mention it.
[05:55] they sometimes just don't mention it. So, you've got this big thing that is
[05:56] So, you've got this big thing that is meant to be the biggest, most
[05:58] meant to be the biggest, most influential change to software ever. And
[06:01] influential change to software ever. And whenever you ask them about it, when you
[06:02] whenever you ask them about it, when you say, "What? How much you making from
[06:03] say, "What? How much you making from this?" They go, "Oh, I couldn't possibly
[06:05] this?" They go, "Oh, I couldn't possibly say. I'm too shy." These are public
[06:07] say. I'm too shy." These are public companies, or at least the ones that
[06:08] companies, or at least the ones that aren't anthropic and open AI. When they
[06:11] aren't anthropic and open AI. When they have good news, they'll tell you. And
[06:12] have good news, they'll tell you. And when they don't tell you something,
[06:13] when they don't tell you something, well, that actually speaks volumes.
[06:16] well, that actually speaks volumes. >> Have you you used these tools, the AI
[06:18] >> Have you you used these tools, the AI tools, Gemini, Anthropic, Chat, GBT,
[06:20] tools, Gemini, Anthropic, Chat, GBT, etc., and you found no value in them?
[06:23] etc., and you found no value in them? There's some value, but it's not there's
[06:25] There's some value, but it's not there's they have spent over a trillion dollars
[06:27] they have spent over a trillion dollars in capex. What
[06:28] in capex. What >> does capex mean for you?
[06:29] >> does capex mean for you? >> Capital expenditures. So, when you are a
[06:31] >> Capital expenditures. So, when you are a business and you have operating expenses
[06:33] business and you have operating expenses like electricity, for example, those
[06:34] like electricity, for example, those come right off immediately. Capital
[06:36] come right off immediately. Capital expenditures are long-term investments
[06:38] expenditures are long-term investments that are theoretically one-off. So, a
[06:39] that are theoretically one-off. So, a data center or indeed the GPUs you put
[06:42] data center or indeed the GPUs you put inside an AI data center.
[06:43] inside an AI data center. >> Okay? So, you've got a data center
[06:45] >> Okay? So, you've got a data center >> and then you have these GPUs which are
[06:47] >> and then you have these GPUs which are like computer chips. So AI GPUs are much
[06:51] like computer chips. So AI GPUs are much bigger, much more power intensive. They
[06:53] bigger, much more power intensive. They take a bunch of high bandwidth memory
[06:55] take a bunch of high bandwidth memory and they because of how many of them you
[06:57] and they because of how many of them you need. You need thousands of them, tens
[06:59] need. You need thousands of them, tens of thousands, hundreds of thousands in
[07:00] of thousands, hundreds of thousands in some case. You need a bunch of power. So
[07:03] some case. You need a bunch of power. So an example, OpenAI and Oracle are
[07:05] an example, OpenAI and Oracle are building a data center in Texas in
[07:07] building a data center in Texas in Abalene, Texas. 1.2 GW called Stargate
[07:10] Abalene, Texas. 1.2 GW called Stargate Abene. Within that, with each one of the
[07:13] Abene. Within that, with each one of the eight buildings, there'll be 50,000
[07:15] eight buildings, there'll be 50,000 Nvidia GB200 GPUs. So, city of Bristol
[07:19] Nvidia GB200 GPUs. So, city of Bristol takes about 7800 megawatt of power a
[07:23] takes about 7800 megawatt of power a year, right? Well, Stargate Abene is
[07:26] year, right? Well, Stargate Abene is condensing more power than that, 1.2
[07:28] condensing more power than that, 1.2 gawatt into a space around 1,172
[07:31] gawatt into a space around 1,172 times smaller. City of Bristol is about
[07:34] times smaller. City of Bristol is about 1.2 billion square ft. Star Evelyn is
[07:37] 1.2 billion square ft. Star Evelyn is about 998,000.
[07:39] about 998,000. So, you're condensing all of this power,
[07:40] So, you're condensing all of this power, all of this money, all of this labor
[07:42] all of this money, all of this labor into this one spot. And all of these
[07:45] into this one spot. And all of these data centers cost billions of dollars.
[07:47] data centers cost billions of dollars. All of these companies other than
[07:48] All of these companies other than Microsoft are now to take out debt. And
[07:50] Microsoft are now to take out debt. And the thing is they've spent over a
[07:52] the thing is they've spent over a trillion dollars so far and they want to
[07:53] trillion dollars so far and they want to spend another trillion dollars next
[07:54] spend another trillion dollars next year. And for what? To make tens of
[07:57] year. And for what? To make tens of billions of dollars, most of which comes
[07:58] billions of dollars, most of which comes from two unprofitable companies,
[08:00] from two unprofitable companies, Anthropic and Open AI. One of the
[08:03] Anthropic and Open AI. One of the rebuttals to that would be that the
[08:04] rebuttals to that would be that the adoption, the customer adoption of
[08:07] adoption, the customer adoption of people using Open AI and Enthropic has
[08:09] people using Open AI and Enthropic has been absolutely insane. These are the
[08:11] been absolutely insane. These are the fastest growing products in all of
[08:13] fastest growing products in all of history, especially as it relates to
[08:14] history, especially as it relates to sort of technology. If we just focus in
[08:16] sort of technology. If we just focus in on technology, they are, you know,
[08:18] on technology, they are, you know, hundreds and hundreds of millions of
[08:19] hundreds and hundreds of millions of people, billions of people are using
[08:21] people, billions of people are using these tools every single day for things
[08:24] these tools every single day for things that they have subjectively decided are
[08:26] that they have subjectively decided are problems they need solving. So, you
[08:29] problems they need solving. So, you know, money is a lagging indicator of
[08:31] know, money is a lagging indicator of value. So, one would argue that they're
[08:34] value. So, one would argue that they're just investing ahead of the monetization
[08:36] just investing ahead of the monetization options.
[08:37] options. >> The first let's start with this
[08:38] >> The first let's start with this adoption. Is it honest adoption when you
[08:41] adoption. Is it honest adoption when you are forced to use generative AI when you
[08:43] are forced to use generative AI when you load Google? When you load Google Docs,
[08:46] load Google? When you load Google Docs, Gemini screams in your ear. When you
[08:48] Gemini screams in your ear. When you load Word, co-pilot's bugging you. When
[08:50] load Word, co-pilot's bugging you. When you use Amazon, whatever rofus AI is
[08:53] you use Amazon, whatever rofus AI is wants has opinions on what socks you're
[08:55] wants has opinions on what socks you're buying. This is the largest
[08:57] buying. This is the largest non-consensual push of technology in
[08:59] non-consensual push of technology in history. Chat GPD for example, every
[09:02] history. Chat GPD for example, every single media outlet has been screaming
[09:03] single media outlet has been screaming about this for 3 years. They've been
[09:06] about this for 3 years. They've been saying, "This will take your job. You
[09:08] saying, "This will take your job. You must use this. If you don't use this,
[09:09] must use this. If you don't use this, you're going to be falling behind." So
[09:11] you're going to be falling behind." So people are using it because they've been
[09:13] people are using it because they've been told to use it constantly and they're
[09:15] told to use it constantly and they're using it like search predominantly and
[09:17] using it like search predominantly and that's partly because Google fell behind
[09:18] that's partly because Google fell behind search and also because it's better at
[09:20] search and also because it's better at ingesting queries sometimes. Sometimes
[09:22] ingesting queries sometimes. Sometimes if you use a generative search it's like
[09:24] if you use a generative search it's like a trolling vessel. It's not very good at
[09:25] a trolling vessel. It's not very good at specifics but if you're like does this
[09:27] specifics but if you're like does this thing exist? Has this person ever said
[09:29] thing exist? Has this person ever said anything like this? It'll still probably
[09:31] anything like this? It'll still probably get it wrong but it'll scour the ocean
[09:33] get it wrong but it'll scour the ocean for you. Nevertheless, that's not worth
[09:35] for you. Nevertheless, that's not worth a trillion dollars. None of it is. The
[09:37] a trillion dollars. None of it is. The amount of money being sunk into this is
[09:40] amount of money being sunk into this is just incomparable to anything. Railways,
[09:43] just incomparable to anything. Railways, it blows everything out of the water
[09:45] it blows everything out of the water because there is no postbubble story
[09:49] because there is no postbubble story even for this. AIG GPU is not useful for
[09:51] even for this. AIG GPU is not useful for other things either. There's it's a
[09:54] other things either. There's it's a directionless egregor of capitalism.
[09:56] directionless egregor of capitalism. this headless beast that lumbers around
[10:00] this headless beast that lumbers around desperate to seek out growth everywhere
[10:02] desperate to seek out growth everywhere in the hopes that if it harasses people
[10:04] in the hopes that if it harasses people and scares people and demonizes labor
[10:08] and scares people and demonizes labor enough, people will be forced to use it.
[10:10] enough, people will be forced to use it. >> The the reason I I pause is because I
[10:13] >> The the reason I I pause is because I just I think about my own company.
[10:14] just I think about my own company. Obviously, everybody thinks about their
[10:15] Obviously, everybody thinks about their own personal situation. So, you have
[10:17] own personal situation. So, you have people listening now that don't use any
[10:18] people listening now that don't use any AI tools. Then you'll have people that
[10:20] AI tools. Then you'll have people that are using it for everything from coding
[10:22] are using it for everything from coding new software tools to everything they
[10:24] new software tools to everything they write to, you know, images, whatever.
[10:26] write to, you know, images, whatever. And when you look at the the stats
[10:28] And when you look at the the stats around enterprise adoption, it says 88%
[10:30] around enterprise adoption, it says 88% of organizations regularly use AI at
[10:32] of organizations regularly use AI at least once for one particular business
[10:34] least once for one particular business function. And I'd say in our company,
[10:37] function. And I'd say in our company, 95% of people use a one of these AI
[10:40] 95% of people use a one of these AI tools like anthropical chatbt or Gemini
[10:42] tools like anthropical chatbt or Gemini every day,
[10:43] every day, >> right? And that exists on some kind of
[10:45] >> right? And that exists on some kind of spectrum of like the super users that
[10:47] spectrum of like the super users that are using it probably, you know, every
[10:50] are using it probably, you know, every hour of every day for almost everything
[10:52] hour of every day for almost everything to, you know, someone maybe hiring the
[10:54] to, you know, someone maybe hiring the executive team that's using it less
[10:56] executive team that's using it less because their job doesn't require of it
[10:57] because their job doesn't require of it as much,
[10:58] as much, >> right?
[10:59] >> right? >> And when you look out into the world,
[11:00] >> And when you look out into the world, you know, at how the world is changing
[11:03] you know, at how the world is changing from a content perspective, if we're
[11:05] from a content perspective, if we're looking at generative AI, it is obvious
[11:07] looking at generative AI, it is obvious that these tools are being widely
[11:09] that these tools are being widely adopted. Part of the symptom is the AI
[11:11] adopted. Part of the symptom is the AI slop you see all over the internet,
[11:12] slop you see all over the internet, >> right?
[11:13] >> right? So, I I don't know this this this idea
[11:16] So, I I don't know this this this idea that it's not being used. I struggle
[11:19] that it's not being used. I struggle with
[11:20] with >> it's being used. Here's the thing with
[11:22] >> it's being used. Here's the thing with the slop. Before we had AI slop, we had
[11:25] the slop. Before we had AI slop, we had SEO slop because Google incentivized
[11:27] SEO slop because Google incentivized doing the lowest common denominator that
[11:29] doing the lowest common denominator that would rank well in search. There's a
[11:31] would rank well in search. There's a whole story about how they pulled back
[11:32] whole story about how they pulled back spam guards thanks to Bravagar Ragavan,
[11:34] spam guards thanks to Bravagar Ragavan, which we can get into,
[11:35] which we can get into, >> where they made the internet worse by
[11:37] >> where they made the internet worse by allowing worse content to rank higher.
[11:39] allowing worse content to rank higher. It's why we have when you used to
[11:41] It's why we have when you used to Google, oh, best washing machine,
[11:43] Google, oh, best washing machine, there's 11 different horrible blogs that
[11:46] there's 11 different horrible blogs that read like somebody got a concussion.
[11:48] read like somebody got a concussion. They are built to rank rather than be
[11:50] They are built to rank rather than be read by humans or built to be good made
[11:52] read by humans or built to be good made good. So AI helps weaponize that at
[11:55] good. So AI helps weaponize that at scale. Yeah, you can make a bunch of
[11:57] scale. Yeah, you can make a bunch of generic slop. We've had slop for years.
[11:59] generic slop. We've had slop for years. We've just found a slop machine. But
[12:01] We've just found a slop machine. But then also there's the problem of cost.
[12:03] then also there's the problem of cost. So when you use AI services, you burn
[12:05] So when you use AI services, you burn tokens and it's per million tokens. So
[12:08] tokens and it's per million tokens. So >> what's a token? So it's around 3/4 of a
[12:10] >> what's a token? So it's around 3/4 of a word. So it's characters.
[12:12] word. So it's characters. >> So the AI companies have a currency in
[12:15] >> So the AI companies have a currency in which they charge you. Like a taxi in
[12:16] which they charge you. Like a taxi in New York has a meter.
[12:17] New York has a meter. >> Yeah.
[12:18] >> Yeah. >> And they call it tokens.
[12:20] >> And they call it tokens. >> Yeah.
[12:21] >> Yeah. >> And every word, let's just say for ease
[12:23] >> And every word, let's just say for ease it's a word. You're paying per word.
[12:24] it's a word. You're paying per word. >> About a word. Yeah. And it's per million
[12:26] >> About a word. Yeah. And it's per million tokens. So you'll be charged per million
[12:28] tokens. So you'll be charged per million input tokens. The stuff you feed into it
[12:30] input tokens. The stuff you feed into it like a document or a bunch a code base.
[12:33] like a document or a bunch a code base. And the output tokens are both the stuff
[12:35] And the output tokens are both the stuff it spits out at the end but also when it
[12:37] it spits out at the end but also when it thinks. So, okay, you've asked me to
[12:39] thinks. So, okay, you've asked me to give you the best restaurants in this
[12:40] give you the best restaurants in this area of New York. I should find the best
[12:42] area of New York. I should find the best restaurants in New York. All of that's
[12:43] restaurants in New York. All of that's output tokens as well.
[12:45] output tokens as well. >> However, when you're paying for a
[12:47] >> However, when you're paying for a monthly service, you don't see any of
[12:48] monthly service, you don't see any of that. Put all that crap to the side.
[12:50] that. Put all that crap to the side. They just have rate limits. So, you can
[12:52] They just have rate limits. So, you can use them a certain amount and then when
[12:53] use them a certain amount and then when you run out, but they kind of offiscate
[12:55] you run out, but they kind of offiscate what that was. Now, someone recently
[12:57] what that was. Now, someone recently found, semi analysis actually found
[12:59] found, semi analysis actually found this, a big analyst group. They found
[13:01] this, a big analyst group. They found that on a $200 a month chat GPD
[13:03] that on a $200 a month chat GPD subscription, you can burn $14,000
[13:06] subscription, you can burn $14,000 worth of tokens and on anthropics you
[13:09] worth of tokens and on anthropics you can burn $8,000 for 200 bucks. That is
[13:14] can burn $8,000 for 200 bucks. That is how most and even on the 20 buck a month
[13:16] how most and even on the 20 buck a month service you can burn $400.
[13:18] service you can burn $400. Now most people don't realize that. Most
[13:20] Now most people don't realize that. Most people have no idea what AI costs. Most
[13:23] people have no idea what AI costs. Most people just think, "Oh, it's 20 bucks a
[13:24] people just think, "Oh, it's 20 bucks a month." No. All of these companies run
[13:26] month." No. All of these companies run at a horrifying loss. OpenAI lost $20.9
[13:30] at a horrifying loss. OpenAI lost $20.9 billion last year because people can
[13:32] billion last year because people can burn as many tokens as they want. And
[13:34] burn as many tokens as they want. And when they tried to move everybody on the
[13:36] when they tried to move everybody on the enterprise side, so companies bigger
[13:38] enterprise side, so companies bigger than 150 onto actually paying the cost
[13:41] than 150 onto actually paying the cost of AI in around March of 2026, to quote
[13:44] of AI in around March of 2026, to quote Sam Orman, they said, uh, people have a
[13:46] Sam Orman, they said, uh, people have a big problem with it. I think it's a huge
[13:47] big problem with it. I think it's a huge issue, which is not really what the air
[13:50] issue, which is not really what the air apparent text history is meant to be
[13:52] apparent text history is meant to be saying, but the point is enterprises
[13:54] saying, but the point is enterprises immediately started freaking out. Uber
[13:57] immediately started freaking out. Uber burned through their entire annual token
[13:58] burned through their entire annual token budget in three months. So suddenly
[14:01] budget in three months. So suddenly after everyone saying AI is the most
[14:03] after everyone saying AI is the most productive thing ever. It's amazing.
[14:04] productive thing ever. It's amazing. It's changing everything. The moment
[14:06] It's changing everything. The moment people actually had to pay for it, they
[14:07] people actually had to pay for it, they go, [snorts]
[14:09] go, [snorts] I don't know actually. Um maybe it's
[14:11] I don't know actually. Um maybe it's obviously we all love it. It's all
[14:13] obviously we all love it. It's all great, right? But it's costing too much.
[14:16] great, right? But it's costing too much. So we need to reduce the cost because
[14:17] So we need to reduce the cost because people are just dumping stuff into it
[14:19] people are just dumping stuff into it being like what do I do here and getting
[14:21] being like what do I do here and getting whatever the median is out because
[14:23] whatever the median is out because that's what these things do. they
[14:25] that's what these things do. they provide the median answer.
[14:27] provide the median answer. >> So essentially, someone like me who's a
[14:28] >> So essentially, someone like me who's a power user of these tools,
[14:30] power user of these tools, >> I could be costing Anthropic or OpenAI
[14:34] >> I could be costing Anthropic or OpenAI $1,000, but they're only charging me
[14:37] $1,000, but they're only charging me $100, let's say. So they are having to
[14:39] $100, let's say. So they are having to subsidize $900 of my usage because of
[14:42] subsidize $900 of my usage because of the electricity costs and the costs at
[14:44] the electricity costs and the costs at their data centers. And so your
[14:45] their data centers. And so your assertion here is that that is
[14:47] assertion here is that that is unsustainable.
[14:48] unsustainable. >> Yes. And just to be clear, they're
[14:49] >> Yes. And just to be clear, they're probably not one for$1. It might be 30
[14:52] probably not one for$1. It might be 30 for. We don't we don't know. I think
[14:53] for. We don't we don't know. I think it's unprofitable. These companies don't
[14:55] it's unprofitable. These companies don't disclose them even in their auditive
[14:56] disclose them even in their auditive financials. They play funny games with
[14:58] financials. They play funny games with how they categorize things. But
[14:59] how they categorize things. But nevertheless, yes. And on top of that,
[15:02] nevertheless, yes. And on top of that, the way that you stand up inference,
[15:04] the way that you stand up inference, which is the thing that creates the
[15:05] which is the thing that creates the output within these data centers, you're
[15:08] output within these data centers, you're not just saying, "Okay, turn the
[15:09] not just saying, "Okay, turn the inference machine on. Let's go." You are
[15:13] inference machine on. Let's go." You are standing up the GPUs necessary to take
[15:15] standing up the GPUs necessary to take in the demand, and if you buy too much,
[15:17] in the demand, and if you buy too much, you've wasted the money. You You have to
[15:19] you've wasted the money. You You have to pay for the hourly GPU use regardless.
[15:22] pay for the hourly GPU use regardless. If you buy too few, your customers can't
[15:24] If you buy too few, your customers can't use it. They get pissed off at you. They
[15:25] use it. They get pissed off at you. They cancel. They go with someone else. But
[15:26] cancel. They go with someone else. But nevertheless, yeah, they would get
[15:28] nevertheless, yeah, they would get demand selling $20 or $40 for a dollar.
[15:31] demand selling $20 or $40 for a dollar. And that's what these services do. And
[15:33] And that's what these services do. And really, the simplest way to explain it
[15:34] really, the simplest way to explain it is they were actually profitable if they
[15:37] is they were actually profitable if they were actually just they believed that
[15:38] were actually just they believed that these services were worthwhile and that
[15:40] these services were worthwhile and that they were worthy of the cost, they'd
[15:42] they were worthy of the cost, they'd charge it. Regular people wouldn't be
[15:43] charge it. Regular people wouldn't be able to get a monthly subscription.
[15:45] able to get a monthly subscription. They'd just be paying what it's worth,
[15:48] They'd just be paying what it's worth, unless, of course, there was an economic
[15:50] unless, of course, there was an economic problem. And it's very simple. You pay
[15:53] problem. And it's very simple. You pay when you use an LLM regardless of
[15:55] when you use an LLM regardless of whether you get what you want. When
[15:57] whether you get what you want. When these things hallucinate, say you're
[15:58] these things hallucinate, say you're doing something, you're coding something
[16:00] doing something, you're coding something and they go through a code base and they
[16:02] and they go through a code base and they up a bunch of stuff, they break a
[16:04] up a bunch of stuff, they break a bunch of stuff, you're paying for that.
[16:05] bunch of stuff, you're paying for that. You're paying for it whether it works or
[16:06] You're paying for it whether it works or not, unless of course you're using one
[16:08] not, unless of course you're using one of these subscriptions. I think the the
[16:10] of these subscriptions. I think the the really interesting point is are they
[16:13] really interesting point is are they spending ahead of the value showing up
[16:17] spending ahead of the value showing up which is I imagine what they would argue
[16:19] which is I imagine what they would argue or are they spending all of this money
[16:22] or are they spending all of this money and subsidizing all of their users in a
[16:25] and subsidizing all of their users in a way that's unsustainable and that will
[16:27] way that's unsustainable and that will never be justified like does it you know
[16:28] never be justified like does it you know because you think back through the
[16:30] because you think back through the history of technology you often get
[16:31] history of technology you often get people
[16:33] people losing money to grab market share
[16:35] losing money to grab market share >> right
[16:35] >> right >> and they're also focusing on bringing
[16:37] >> and they're also focusing on bringing the costs down and making it more
[16:40] the costs down and making it more profitable for them as well. But they
[16:42] profitable for them as well. But they can't afford to underinvest.
[16:44] can't afford to underinvest. >> If they were bringing the cost down,
[16:46] >> If they were bringing the cost down, they would have brought the cost down,
[16:48] they would have brought the cost down, which they have not. It seems to be
[16:50] which they have not. It seems to be getting more expensive. In fact,
[16:51] getting more expensive. In fact, everyone inference providers don't seem
[16:53] everyone inference providers don't seem to be profitable. Even the companies
[16:55] to be profitable. Even the companies renting out GPUs don't seem to be
[16:56] renting out GPUs don't seem to be profitable. I imagine that it wasn't
[16:59] profitable. I imagine that it wasn't like they started out and they were
[17:00] like they started out and they were like, "Shit, this is unprofitable at the
[17:02] like, "Shit, this is unprofitable at the beginning. We know it. Screw it. We'll
[17:04] beginning. We know it. Screw it. We'll keep doing it any screw." I don't think
[17:05] keep doing it any screw." I don't think it's some big conspiracy. They probably
[17:07] it's some big conspiracy. They probably thought at some point, yeah, this will
[17:10] thought at some point, yeah, this will go profitable. The chips will catch up.
[17:12] go profitable. The chips will catch up. Customers will pay for the overwhelming
[17:14] Customers will pay for the overwhelming value because you don't know in 2023
[17:15] value because you don't know in 2023 where it's going to be in 2026. You
[17:17] where it's going to be in 2026. You assume it's going to go up. That's the
[17:18] assume it's going to go up. That's the nature of venture capital. They should
[17:20] nature of venture capital. They should have stopped in like 2024 when OpenAI
[17:23] have stopped in like 2024 when OpenAI lost over $5 billion. They should have
[17:25] lost over $5 billion. They should have been like, "Yep, this is not going to
[17:26] been like, "Yep, this is not going to work." But they kept going because it
[17:29] work." But they kept going because it helped number go up so much. It helped
[17:31] helped number go up so much. It helped stock values pump. It helped everyone
[17:33] stock values pump. It helped everyone pump. It helped Nvidia pump, Microsoft,
[17:35] pump. It helped Nvidia pump, Microsoft, everyone. and not from the revenues.
[17:38] everyone. and not from the revenues. Because here's the funny thing about
[17:39] Because here's the funny thing about Google, Microsoft, and Amazon. People
[17:41] Google, Microsoft, and Amazon. People for years have been saying their AI bets
[17:44] for years have been saying their AI bets have paid off. Wow, their AI bets have
[17:46] have paid off. Wow, their AI bets have paid off. As these companies refused to
[17:47] paid off. As these companies refused to say how much they're making from AI, but
[17:49] say how much they're making from AI, but because their existing businesses
[17:51] because their existing businesses continued to grow and did so, by the
[17:53] continued to grow and did so, by the way, through price increases, changes to
[17:55] way, through price increases, changes to how Google and Meta uh did advertising.
[17:58] how Google and Meta uh did advertising. Amazon bumped up prices and changed how
[18:00] Amazon bumped up prices and changed how they did actually Amazon started a
[18:02] they did actually Amazon started a remarkable ad business during this whole
[18:04] remarkable ad business during this whole time as well. and the selling through
[18:05] time as well. and the selling through Amazon platform anyway nothing to do
[18:07] Amazon platform anyway nothing to do with AI but because number go up because
[18:09] with AI but because number go up because revenue go up everyone went it's AI
[18:11] revenue go up everyone went it's AI because these companies wouldn't spend a
[18:12] because these companies wouldn't spend a trillion dollars for for no reason right
[18:16] trillion dollars for for no reason right except in fiscal year 2026 which just
[18:18] except in fiscal year 2026 which just ended for Microsoft annoying I know they
[18:21] ended for Microsoft annoying I know they made total according to Bloomberg about
[18:23] made total according to Bloomberg about $34.33 billion $24.1 billion of that was
[18:28] $34.33 billion $24.1 billion of that was from OpenAI so that leaves them with
[18:30] from OpenAI so that leaves them with about $10 billion in a year when they
[18:32] about $10 billion in a year when they spent 115 billion on capital
[18:34] spent 115 billion on capital expenditures just intend to spend 175
[18:36] expenditures just intend to spend 175 billion next year. The math does not
[18:39] billion next year. The math does not make sense. I imagine their plan was
[18:41] make sense. I imagine their plan was okay, this is just going to get
[18:42] okay, this is just going to get exponentially more valuable and at some
[18:44] exponentially more valuable and at some point the costs will be outpaced by the
[18:46] point the costs will be outpaced by the return. Problem is that large language
[18:49] return. Problem is that large language models need a bunch of money to train
[18:50] models need a bunch of money to train them. They need constant data flow. They
[18:53] them. They need constant data flow. They need customized data. It's just this big
[18:55] need customized data. It's just this big expensive monster. And when you try and
[18:59] expensive monster. And when you try and talk to people about it and you try and
[19:00] talk to people about it and you try and say, "Hey, look, this is really bad.
[19:02] say, "Hey, look, this is really bad. Nvidia has sold it was $215.9 billion in
[19:07] Nvidia has sold it was $215.9 billion in the last fiscal year worth of GPUs
[19:09] the last fiscal year worth of GPUs mostly. And you try and go, yeah, that's
[19:11] mostly. And you try and go, yeah, that's to support like $22 billion of revenue
[19:15] to support like $22 billion of revenue total in the entire world outside of
[19:18] total in the entire world outside of these two companies that literally
[19:20] these two companies that literally require money being fed into them
[19:21] require money being fed into them sometimes by Nvidia to keep alive. When
[19:24] sometimes by Nvidia to keep alive. When you tell people that, they go, "Well,
[19:26] you tell people that, they go, "Well, companies just lose money, right?
[19:27] companies just lose money, right? Companies because we have this quote
[19:29] Companies because we have this quote Edson from Prophy Markets. We have this
[19:31] Edson from Prophy Markets. We have this cult-like worship of the wealthy where
[19:33] cult-like worship of the wealthy where we think that someone wouldn't spend all
[19:35] we think that someone wouldn't spend all this money for no reason. Right? Because
[19:37] this money for no reason. Right? Because reconciling with that with this idea
[19:39] reconciling with that with this idea that the ultra wealthy, the ultra
[19:43] that the ultra wealthy, the ultra powerful didn't get there through big
[19:45] powerful didn't get there through big brains. They didn't get there through
[19:47] brains. They didn't get there through anything other than luck and opportunism
[19:50] anything other than luck and opportunism and getting an MBA perhaps with the
[19:52] and getting an MBA perhaps with the right people. That they just got there
[19:54] right people. That they just got there because they're regular people and they
[19:56] because they're regular people and they just happen to be in the right place at
[19:57] just happen to be in the right place at the right time. reconciling with that
[19:58] the right time. reconciling with that and realizing that the world is not
[19:59] and realizing that the world is not controlled by people like a meritocracy
[20:01] controlled by people like a meritocracy is kind of grim. So it's easy to be like
[20:04] is kind of grim. So it's easy to be like no they're not making a mistake I must
[20:06] no they're not making a mistake I must be missing something and that's what
[20:07] be missing something and that's what they want. So you know I think back
[20:09] they want. So you know I think back through the history of technological
[20:11] through the history of technological breakthroughs and I think about I mean
[20:13] breakthroughs and I think about I mean you can look at different industries and
[20:14] you can look at different industries and one of my favorite books on this subject
[20:16] one of my favorite books on this subject is the innovator's dilemma. not read it.
[20:18] is the innovator's dilemma. not read it. >> And one of the things it talks about is
[20:20] >> And one of the things it talks about is how the the innovation that ends up
[20:22] how the the innovation that ends up taking out or transforming an industry
[20:25] taking out or transforming an industry often starts worse, doesn't make
[20:28] often starts worse, doesn't make economic sense, none of your customers
[20:30] economic sense, none of your customers are asking for it. And this is typically
[20:32] are asking for it. And this is typically why we end up ignoring it. So like
[20:34] why we end up ignoring it. So like you've got horse and carriages in the
[20:35] you've got horse and carriages in the 1800s.
[20:36] 1800s. >> Amazing form of transport according to
[20:38] >> Amazing form of transport according to the 1800s, you know, people of the
[20:39] the 1800s, you know, people of the 1800s. And then you have this thing
[20:41] 1800s. And then you have this thing called cars come along. Now the problem
[20:42] called cars come along. Now the problem with cars is they broke down all the
[20:44] with cars is they broke down all the time. It's kind of like AI hallucinates
[20:45] time. It's kind of like AI hallucinates now. um they were more expensive and the
[20:48] now. um they were more expensive and the the economics of it didn't make sense.
[20:49] the economics of it didn't make sense. You might as well walk than buy a car.
[20:51] You might as well walk than buy a car. There was a law at the time that meant
[20:52] There was a law at the time that meant you had to walk in front of it with a
[20:53] you had to walk in front of it with a red flag and wave and someone had you
[20:55] red flag and wave and someone had you had to employ someone to walk in front
[20:56] had to employ someone to walk in front of it waving a red flag. Obviously, it's
[20:59] of it waving a red flag. Obviously, it's worse. It's like a worse solution.
[21:00] worse. It's like a worse solution. However, these things that are
[21:02] However, these things that are disruptive innovations, they have a
[21:05] disruptive innovations, they have a higher ceiling of growth and so they
[21:07] higher ceiling of growth and so they eventually overtake the horse. And I
[21:10] eventually overtake the horse. And I when I think about that analogy in the
[21:11] when I think about that analogy in the context of all of this, I go, okay, it's
[21:13] context of all of this, I go, okay, it's imperfect at the at the moment. the
[21:16] imperfect at the at the moment. the economic models aren't perfectly ironed
[21:18] economic models aren't perfectly ironed out. They're still figuring out how to
[21:20] out. They're still figuring out how to make it cheaper, the infrastructure,
[21:21] make it cheaper, the infrastructure, etc. But as if you think about the rate
[21:23] etc. But as if you think about the rate of improvement versus other you know
[21:26] of improvement versus other you know let's say coding how much could I train
[21:29] let's say coding how much could I train a human coder to improve and to increase
[21:31] a human coder to improve and to increase their output versus an AI agent one
[21:33] their output versus an AI agent one would go if you just imagine any rate of
[21:36] would go if you just imagine any rate of improvement in these AI tools at some
[21:38] improvement in these AI tools at some point if you just imagine a 5% rate of
[21:40] point if you just imagine a 5% rate of improvement per month at some point it's
[21:44] improvement per month at some point it's you know and then you imagine a 5%
[21:45] you know and then you imagine a 5% reduction in cost which is what we did
[21:47] reduction in cost which is what we did with the internet what we did with cars
[21:49] with the internet what we did with cars but Mo's law
[21:50] but Mo's law >> mos law is a mos law is not with GPUs.
[21:52] >> mos law is a mos law is not with GPUs. So let me let me actually explain. So
[21:54] So let me let me actually explain. So Nvidia Nvidia invented I think it was in
[21:56] Nvidia Nvidia invented I think it was in the 2000s they put out something called
[21:58] the 2000s they put out something called CUDA which is the underlying software
[22:00] CUDA which is the underlying software library and the way to run software on
[22:03] library and the way to run software on GPUs. took them solid decade or more to
[22:06] GPUs. took them solid decade or more to make it something where they could do
[22:08] make it something where they could do data analytics, one of the early things,
[22:10] data analytics, one of the early things, mapper and such. And then when AI came
[22:12] mapper and such. And then when AI came along, they'd had lots of experience
[22:13] along, they'd had lots of experience with it. But nevertheless, this company
[22:15] with it. But nevertheless, this company has got more money, more attention, more
[22:19] has got more money, more attention, more geniuses behind them, more people
[22:21] geniuses behind them, more people focused on making their things more
[22:22] focused on making their things more efficient than anyone could ever ask
[22:25] efficient than anyone could ever ask for.
[22:25] for. >> And Nvidia, for anyone that doesn't
[22:26] >> And Nvidia, for anyone that doesn't know, makes the chips.
[22:27] know, makes the chips. >> They So, and that CUDA thing I
[22:29] >> They So, and that CUDA thing I mentioned, they were the ones with CUDA
[22:31] mentioned, they were the ones with CUDA and CUDA allowed generative AI to grow.
[22:33] and CUDA allowed generative AI to grow. Okay, so they're chips.
[22:34] Okay, so they're chips. >> Chips and chips are needed. Those are
[22:36] >> Chips and chips are needed. Those are the things that go into the data
[22:37] the things that go into the data centers.
[22:37] centers. >> And there specific chips are the ones
[22:39] >> And there specific chips are the ones where you can run AI software on it. So
[22:41] where you can run AI software on it. So the training runs and also the
[22:42] the training runs and also the inference. Now, here's the thing. The
[22:44] inference. Now, here's the thing. The the car example back then you didn't
[22:47] the car example back then you didn't have pretty much every mathematician and
[22:49] have pretty much every mathematician and scientist going into the car industry.
[22:51] scientist going into the car industry. You didn't have the combined world's
[22:53] You didn't have the combined world's governments never shutting up about
[22:54] governments never shutting up about this. And by the way, giving them credit
[22:58] this. And by the way, giving them credit early since 2023, they've been saying
[23:00] early since 2023, they've been saying this is inevitable. Even in what you
[23:02] this is inevitable. Even in what you said, 5% improvement. I don't even know
[23:04] said, 5% improvement. I don't even know how you'd measure that because a junior
[23:06] how you'd measure that because a junior software engineer can still experience
[23:08] software engineer can still experience things and learn things from context,
[23:10] things and learn things from context, from how people deal with problems. And
[23:12] from how people deal with problems. And the way that people deal with problems
[23:14] the way that people deal with problems is not as simple as looking at the code
[23:16] is not as simple as looking at the code or reading some emails. It's context
[23:18] or reading some emails. It's context cues from speaking to a person. It's
[23:19] cues from speaking to a person. It's being in different environments. And
[23:22] being in different environments. And there may there are uses for LLM's
[23:24] there may there are uses for LLM's encoding. I don't dispute that. But even
[23:26] encoding. I don't dispute that. But even saying 5% uh what does that mean? Is it
[23:29] saying 5% uh what does that mean? Is it better at Rust? Is it better at C++?
[23:31] better at Rust? Is it better at C++? >> I'd say productivity just like yeah
[23:33] >> I'd say productivity just like yeah shipped. If we did it in the context of
[23:34] shipped. If we did it in the context of coding, it would be like shipped code.
[23:36] coding, it would be like shipped code. >> That's the thing that would be like he's
[23:38] >> That's the thing that would be like he's the best writer in the world cuz his
[23:40] the best writer in the world cuz his newsletter's really long. That's an
[23:41] newsletter's really long. That's an insane way of evaluing it. With coding,
[23:44] insane way of evaluing it. With coding, it would be I mean it's even difficult
[23:46] it would be I mean it's even difficult to evaluate because it's is the software
[23:48] to evaluate because it's is the software out there better is actually a great way
[23:50] out there better is actually a great way of evaluating it. And I would say
[23:52] of evaluating it. And I would say uniformly not. I would say the standard
[23:54] uniformly not. I would say the standard of software across Google, Microsoft,
[23:56] of software across Google, Microsoft, Amazon, Meta, especially God, Meta is a
[23:59] Amazon, Meta, especially God, Meta is a monstrosity, is worse. GitHub, GitHub,
[24:03] monstrosity, is worse. GitHub, GitHub, someone posted on Twitter earlier today,
[24:04] someone posted on Twitter earlier today, we should get a notification when GitHub
[24:06] we should get a notification when GitHub is up rather than when it's down because
[24:08] is up rather than when it's down because that would be more reliable. Microsoft's
[24:09] that would be more reliable. Microsoft's one of the largest companies in the
[24:10] one of the largest companies in the world, and they can barely wipe their
[24:11] world, and they can barely wipe their own ass when it comes to GitHub. The
[24:14] own ass when it comes to GitHub. The quality of software is going down
[24:16] quality of software is going down weirdly enough as more people use LLMs
[24:18] weirdly enough as more people use LLMs and more businesses demand and I really
[24:21] and more businesses demand and I really do mean demand that people use these
[24:23] do mean demand that people use these services. So on this point of if we go
[24:25] services. So on this point of if we go back to this horse and carriage and car
[24:26] back to this horse and carriage and car analogy say that we're at whatever point
[24:29] analogy say that we're at whatever point today if you imagine any rate of
[24:31] today if you imagine any rate of improvement in the technology which we
[24:33] improvement in the technology which we have seen since tragedy came out
[24:35] have seen since tragedy came out >> I remember when tragy came out and I was
[24:36] >> I remember when tragy came out and I was in Asia and I was there showing it to my
[24:38] in Asia and I was there showing it to my fiance I was like look it can do this
[24:39] fiance I was like look it can do this and it was hallucinating once in a while
[24:41] and it was hallucinating once in a while and getting things wrong. I actually
[24:43] and getting things wrong. I actually don't have that experience anymore. I
[24:45] don't have that experience anymore. I have moments where I believe it's
[24:47] have moments where I believe it's reasoning is weak, but I don't have
[24:50] reasoning is weak, but I don't have outright hallucinations anymore. See
[24:52] outright hallucinations anymore. See that? I I disagree. So,
[24:54] that? I I disagree. So, >> give me an example of what you define as
[24:56] >> give me an example of what you define as a hallucination.
[24:56] a hallucination. >> Okay, great one. So, I have a Bloomberg
[24:58] >> Okay, great one. So, I have a Bloomberg terminal. Yeah. The very useful thing
[25:00] terminal. Yeah. The very useful thing they have on there is ask B. So, when
[25:01] they have on there is ask B. So, when you do a Bloomberg inquiry to like look
[25:03] you do a Bloomberg inquiry to like look up what we think Nvidia's revenue is
[25:05] up what we think Nvidia's revenue is going to be next quarter, it runs
[25:07] going to be next quarter, it runs something called BQL, which is its own
[25:09] something called BQL, which is its own programming language. Now, instead of
[25:11] programming language. Now, instead of having to learn that, you can just type
[25:13] having to learn that, you can just type into RSB and it will generate it and run
[25:15] into RSB and it will generate it and run it for you. And so, you get it pulled up
[25:16] it for you. And so, you get it pulled up and you know where the data is coming
[25:17] and you know where the data is coming from. It deals with hallucinations real
[25:19] from. It deals with hallucinations real well. The other day, I was like, you
[25:21] well. The other day, I was like, you know what, get a little spicy. I'm going
[25:22] know what, get a little spicy. I'm going to look up the growth rate of stocks of
[25:26] to look up the growth rate of stocks of Microsoft, Google, Meta, and Amazon over
[25:28] Microsoft, Google, Meta, and Amazon over the course of 5 years, I think it was.
[25:30] the course of 5 years, I think it was. >> And I was about to I was copy pasted it
[25:33] >> And I was about to I was copy pasted it over to something looked at in Excel. I
[25:34] over to something looked at in Excel. I was about to was writing the newsletter.
[25:36] was about to was writing the newsletter. I went, Microsoft stocks never been $575
[25:39] I went, Microsoft stocks never been $575 a stock.
[25:41] a stock. You know what? When it's a cute little
[25:43] You know what? When it's a cute little thing like, oh, it's a stock price and I
[25:44] thing like, oh, it's a stock price and I kind of call it was no harm, no foul.
[25:46] kind of call it was no harm, no foul. That's fine. But when you're talking
[25:48] That's fine. But when you're talking about, I don't know, like a transcribing
[25:51] about, I don't know, like a transcribing tool for a doctor or a financial model
[25:53] tool for a doctor or a financial model that a hedge fund is dependent on, at
[25:56] that a hedge fund is dependent on, at that point it becomes a little more
[25:58] that point it becomes a little more dangerous. And the thing is a
[26:00] dangerous. And the thing is a hallucination with a software package.
[26:03] hallucination with a software package. For example, you're refactoring a code
[26:05] For example, you're refactoring a code base and it leaves a door open
[26:07] base and it leaves a door open security-wise or it just breaks
[26:09] security-wise or it just breaks something and you I don't know maybe
[26:11] something and you I don't know maybe you've been vibe coding for 6 months.
[26:12] you've been vibe coding for 6 months. You haven't really been coding with your
[26:14] You haven't really been coding with your own hands for a while. Maybe you've
[26:15] own hands for a while. Maybe you've forgotten a few things. You had this
[26:17] forgotten a few things. You had this slop to look for. I'm not doing
[26:19] slop to look for. I'm not doing it. And so the problems become
[26:21] it. And so the problems become multiplicative. And I don't really know
[26:24] multiplicative. And I don't really know how you train them out of that. And
[26:26] how you train them out of that. And they've certainly not succeeded. So on
[26:28] they've certainly not succeeded. So on one hand they have got better but one of
[26:31] one hand they have got better but one of the main ways they evaluate them getting
[26:33] the main ways they evaluate them getting better are benchmarks that are adjusted
[26:35] better are benchmarks that are adjusted specifically for large language models
[26:37] specifically for large language models because you can't just have them do
[26:39] because you can't just have them do tasks. They've got better at that. They
[26:41] tasks. They've got better at that. They found some tasks they can have them do
[26:43] found some tasks they can have them do on them like meter me they have this
[26:46] on them like meter me they have this thing where it's like check out this
[26:47] thing where it's like check out this chart look how much better it's getting
[26:48] chart look how much better it's getting at running tasks. Wow it can go for an
[26:50] at running tasks. Wow it can go for an hour and then you look it's like yeah
[26:52] hour and then you look it's like yeah and successfully completing them 50% of
[26:54] and successfully completing them 50% of the time. They they have a hallucination
[26:55] the time. They they have a hallucination leaderboard and it really focuses on
[26:57] leaderboard and it really focuses on basic tasks and it shows that the
[26:59] basic tasks and it shows that the four-year trend according to historical
[27:00] four-year trend according to historical data from the Victaria hallucination
[27:03] data from the Victaria hallucination leaderboard shows that hallucination
[27:06] leaderboard shows that hallucination rates on simple summarization tasks have
[27:08] rates on simple summarization tasks have plummeted from around 21% 21.8% 4 years
[27:11] plummeted from around 21% 21.8% 4 years ago down to 0.7%
[27:15] ago down to 0.7% roughly on today's top frontier models
[27:17] roughly on today's top frontier models like Gemini and Chat GPT. Again the
[27:20] like Gemini and Chat GPT. Again the point of nuance here is that these are
[27:21] point of nuance here is that these are on simple tasks which is kind of what
[27:24] on simple tasks which is kind of what I've experienced. I've experienced that
[27:25] I've experienced. I've experienced that on day-to-day things that hallucinates
[27:26] on day-to-day things that hallucinates less again rate of improvement thinking.
[27:28] less again rate of improvement thinking. So if I just imagine the trajectory to
[27:31] So if I just imagine the trajectory to continue there is going to become a time
[27:33] continue there is going to become a time where hallucinations become rarer than
[27:36] where hallucinations become rarer than they are today increasingly and also
[27:38] they are today increasingly and also what I would say is when I think about
[27:39] what I would say is when I think about other technologies there's two more
[27:41] other technologies there's two more points other technologies at their
[27:42] points other technologies at their inception when they first came to the
[27:44] inception when they first came to the world like the internet also had
[27:46] world like the internet also had technical difficulties. I remember
[27:48] technical difficulties. I remember growing up with dialup modems and I
[27:50] growing up with dialup modems and I couldn't go on the phone at the same
[27:51] couldn't go on the phone at the same time as going on the internet. I'd have
[27:52] time as going on the internet. I'd have to stop Runescape upstairs to go on the
[27:55] to stop Runescape upstairs to go on the phone. And you thought this is crap.
[27:56] phone. And you thought this is crap. This is technology crap. All the
[27:58] This is technology crap. All the >> I I don't know, mate. I loved it.
[27:59] >> I I don't know, mate. I loved it. >> Yeah, I know. You It felt like magic.
[28:01] >> Yeah, I know. You It felt like magic. And then in hindsight, you go, "Wow, I
[28:03] And then in hindsight, you go, "Wow, I now have Starink and 5G internet from my
[28:06] now have Starink and 5G internet from my phone. It's unbelievable." You couldn't
[28:08] phone. It's unbelievable." You couldn't leave the house with internet before.
[28:10] leave the house with internet before. And that's what I mean by the rate of
[28:11] And that's what I mean by the rate of improvement thinking. I'd say the last
[28:12] improvement thinking. I'd say the last point is we often compare AI to
[28:17] point is we often compare AI to perfection,
[28:18] perfection, >> right?
[28:18] >> right? >> Whereas that's not actually the
[28:19] >> Whereas that's not actually the alternative in the working world. Like
[28:22] alternative in the working world. Like if I wanted to do let's say a simple
[28:24] if I wanted to do let's say a simple writing task, I should compare AI to my
[28:28] writing task, I should compare AI to my alternative alternative way of doing
[28:30] alternative alternative way of doing that simple writing task which is both
[28:31] that simple writing task which is both measured in my time right and my ability
[28:34] measured in my time right and my ability to hallucinate as a person who doesn't
[28:36] to hallucinate as a person who doesn't know everything
[28:38] know everything or if I'm hiring someone an intern who
[28:41] or if I'm hiring someone an intern who might also be prone to hallucination or
[28:43] might also be prone to hallucination or have gaps in their knowledge.
[28:44] have gaps in their knowledge. >> So it's not actually like we're
[28:46] >> So it's not actually like we're comparing we should compare AI to
[28:47] comparing we should compare AI to perfection. It's AI to the other
[28:48] perfection. It's AI to the other alternatives. And if someone
[28:50] alternatives. And if someone hallucinates 0.7% of the time, but knows
[28:53] hallucinates 0.7% of the time, but knows way more and is faster, maybe on a net
[28:56] way more and is faster, maybe on a net basis, that's a good trade. Maybe I
[28:58] basis, that's a good trade. Maybe I should use AI. So, let's start with an
[29:01] should use AI. So, let's start with an example. Someone I love dearly, Matt
[29:02] example. Someone I love dearly, Matt Hughes, my editor, lives out of
[29:04] Hughes, my editor, lives out of Liverpool. Wonderful guy. I don't pay
[29:07] Liverpool. Wonderful guy. I don't pay Matt Hughes because he knows everything.
[29:09] Matt Hughes because he knows everything. I pay him because he has incredible
[29:11] I pay him because he has incredible context and a ton of knowledge and he's
[29:14] context and a ton of knowledge and he's willing to expand it and work with me
[29:16] willing to expand it and work with me and moral sport and he's a great editor,
[29:18] and moral sport and he's a great editor, but he's also someone who gets into the
[29:20] but he's also someone who gets into the guts of it and has the experiences of
[29:22] guts of it and has the experiences of it. He's a decorated tech journalist and
[29:24] it. He's a decorated tech journalist and on top of that a wonderful loving being
[29:27] on top of that a wonderful loving being with empathy and joy in his heart for
[29:29] with empathy and joy in his heart for the stuff he loves and absolute
[29:31] the stuff he loves and absolute venom for the people he hates. That's I
[29:33] venom for the people he hates. That's I can't get that from a large language
[29:34] can't get that from a large language model. But on top of that, I don't I
[29:36] model. But on top of that, I don't I push back on just the assumption there.
[29:39] push back on just the assumption there. >> When you say knows everything, what good
[29:41] >> When you say knows everything, what good is something that knows everything when
[29:42] is something that knows everything when it sometimes doesn't know anything when
[29:44] it sometimes doesn't know anything when it's sometimes? And on the thing is, are
[29:46] it's sometimes? And on the thing is, are you really paying an intern for
[29:48] you really paying an intern for something basic? Are you really going to
[29:50] something basic? Are you really going to them and saying, "Yeah, can you look up
[29:52] them and saying, "Yeah, can you look up what the date is?" No, you're doing that
[29:53] what the date is?" No, you're doing that on Google. Whatever the task is, you are
[29:56] on Google. Whatever the task is, you are trying to also train an intern. The
[29:58] trying to also train an intern. The point of an intern is to train them and
[29:59] point of an intern is to train them and turn them in, take them out of Pinocchio
[30:01] turn them in, take them out of Pinocchio status,
[30:02] status, >> but it's also an intern learns. And in
[30:04] >> but it's also an intern learns. And in turn gets context and in turn learns
[30:06] turn gets context and in turn learns your habits. Learns
[30:07] your habits. Learns >> AI gets context and learns.
[30:08] >> AI gets context and learns. >> No, it doesn't. It
[30:09] >> No, it doesn't. It >> doesn't learn.
[30:10] >> doesn't learn. >> I mean, it doesn't. The way it learns is
[30:12] >> I mean, it doesn't. The way it learns is you create a giant claw. MD file that it
[30:15] you create a giant claw. MD file that it sometimes doesn't read, sometimes does
[30:16] sometimes doesn't read, sometimes does read. You create a harness. You put it's
[30:18] read. You create a harness. You put it's like it's Pee-Wee's breakfast machine
[30:20] like it's Pee-Wee's breakfast machine from PeeWee's Playhouse. You have to do
[30:22] from PeeWee's Playhouse. You have to do all these controversies to mitigate the
[30:25] all these controversies to mitigate the hallucinations. And even then at the
[30:26] hallucinations. And even then at the end, how much effort have you put in?
[30:28] end, how much effort have you put in? >> But so, okay, this is an extreme
[30:30] >> But so, okay, this is an extreme simplified example. If I went on my
[30:32] simplified example. If I went on my Claude now and said, "What's my dog? my
[30:33] Claude now and said, "What's my dog? my dog's name.
[30:34] dog's name. >> Uhhuh.
[30:34] >> Uhhuh. >> It would know my dog's name.
[30:35] >> It would know my dog's name. >> Jesus Christ. This this company raised
[30:37] >> Jesus Christ. This this company raised 95 billion.
[30:39] 95 billion. >> I'm saying I'm I'm using an extreme
[30:40] >> I'm saying I'm I'm using an extreme simplified example to show that it can
[30:42] simplified example to show that it can remember things from the past.
[30:43] remember things from the past. Obviously, it knows much more complex
[30:45] Obviously, it knows much more complex things as well, but I just use that as
[30:47] things as well, but I just use that as an example. So, we we we accept the fact
[30:49] an example. So, we we we accept the fact that it can it does have memory of the
[30:51] that it can it does have memory of the past.
[30:51] past. >> It has files it can access that have
[30:53] >> It has files it can access that have stuff on it, but that's not the same as
[30:55] stuff on it, but that's not the same as memory. And it's also just okay. So, it
[30:58] memory. And it's also just okay. So, it remembers your dog's name. It might
[30:59] remembers your dog's name. It might remember your habits. It might be able
[31:01] remember your habits. It might be able to read things you've said before.
[31:03] to read things you've said before. >> Does it know your moods? Does it know
[31:05] >> Does it know your moods? Does it know what's going on in the world around it?
[31:07] what's going on in the world around it? Does it have good days and bad days? Is
[31:08] Does it have good days and bad days? Is it there for you? Because it's just a
[31:10] it there for you? Because it's just a text machine. And the thing is
[31:13] text machine. And the thing is the intern example. An intern is
[31:15] the intern example. An intern is something that can grow. It's something
[31:17] something that can grow. It's something that you invest in. That's not something
[31:19] that you invest in. That's not something you do through feeding files and text to
[31:20] you do through feeding files and text to it. The way that we store memories
[31:23] it. The way that we store memories ourselves, the way in which we acrue
[31:25] ourselves, the way in which we acrue experiences is a a milerum of emotion
[31:29] experiences is a a milerum of emotion and feelings and facts
[31:31] and feelings and facts >> completely different. So I think there's
[31:33] >> completely different. So I think there's two things here. There's the process in
[31:34] two things here. There's the process in which something happens and then there's
[31:36] which something happens and then there's the output.
[31:37] the output. >> So the process you're describing the
[31:39] >> So the process you're describing the process of how a human does memory,
[31:41] process of how a human does memory, >> right?
[31:41] >> right? >> The way that an AI does memory is
[31:43] >> The way that an AI does memory is different. But the thing that people
[31:45] different. But the thing that people care about is there value in the output.
[31:48] care about is there value in the output. I.e. You know, if I dump all of my files
[31:50] I.e. You know, if I dump all of my files into Claude, I don't really care how it
[31:52] into Claude, I don't really care how it processes it as long as when I ask it,
[31:55] processes it as long as when I ask it, what's my revenue? It has the number.
[31:57] what's my revenue? It has the number. And one could say the same thing about
[31:58] And one could say the same thing about training someone. You could say, you
[32:00] training someone. You could say, you teach them, you put lots of effort into
[32:01] teach them, you put lots of effort into them. You give them lots of context. You
[32:03] them. You give them lots of context. You you educate them and give them
[32:04] you educate them and give them experiences. And then you might come and
[32:06] experiences. And then you might come and say to them, by the way, what's my
[32:07] say to them, by the way, what's my revenue? Now, the processes are entirely
[32:09] revenue? Now, the processes are entirely different, but the outcome is what I
[32:10] different, but the outcome is what I care about. Do they know the revenue
[32:12] care about. Do they know the revenue number when I ask them? And so, I think
[32:14] number when I ask them? And so, I think that's the part that we sometimes get
[32:15] that's the part that we sometimes get lost. we get, you know, cuz I have I've
[32:17] lost. we get, you know, cuz I have I've heard this debate about like can AI be
[32:19] heard this debate about like can AI be creative,
[32:19] creative, >> right?
[32:20] >> right? >> I think like the way to answer that
[32:21] >> I think like the way to answer that question is like it's about the output
[32:24] question is like it's about the output when I ask it to do a creative thing
[32:25] when I ask it to do a creative thing does it give me the answer not is the
[32:28] does it give me the answer not is the process the same as a human process cuz
[32:30] process the same as a human process cuz actually no who cares what the people
[32:32] actually no who cares what the people care about they pay for the outcome the
[32:34] care about they pay for the outcome the product.
[32:35] product. >> I actually disagree about the process
[32:37] >> I actually disagree about the process because Matt Hughes for example
[32:40] because Matt Hughes for example >> your editor
[32:40] >> your editor >> Yeah.
[32:41] >> Yeah. >> Yeah. watching him go down a rabbit hole
[32:43] >> Yeah. watching him go down a rabbit hole and being there with him and actually
[32:45] and being there with him and actually vice versa him doing the same thing. We
[32:47] vice versa him doing the same thing. We wrote these well I mean we were working
[32:49] wrote these well I mean we were working on the research I ended up sitting there
[32:51] on the research I ended up sitting there for like the dayong session of writing
[32:53] for like the dayong session of writing 11,000 words and he he had given me a
[32:55] 11,000 words and he he had given me a bunch of notes. It was actually just
[32:57] bunch of notes. It was actually just even describing that process, I feel so
[32:59] even describing that process, I feel so happy cuz it was like us being like I
[33:01] happy cuz it was like us being like I can't believe how these Jesus
[33:02] can't believe how these Jesus Christ they can't do like just like the
[33:05] Christ they can't do like just like the misanthropy of just the horrible cynical
[33:07] misanthropy of just the horrible cynical people of asset managers like Blackstone
[33:10] people of asset managers like Blackstone just learning about them and being like
[33:11] just learning about them and being like it can't be this and having a back and
[33:12] it can't be this and having a back and forth with him that is fundamentally
[33:14] forth with him that is fundamentally different because we were both learning
[33:16] different because we were both learning together and the learning process was as
[33:18] together and the learning process was as much about creating the output as the
[33:20] much about creating the output as the output itself. When you learn something,
[33:22] output itself. When you learn something, you're not creating the average, which
[33:24] you're not creating the average, which really is what these things do, of the
[33:26] really is what these things do, of the documents it could find. You're not
[33:28] documents it could find. You're not getting particularly novel outputs. If I
[33:31] getting particularly novel outputs. If I needed a generic slop output, sure, but
[33:34] needed a generic slop output, sure, but I've I've used some of the higherend LLM
[33:37] I've I've used some of the higherend LLM harness machines that the hedge funds
[33:40] harness machines that the hedge funds use, and they all give the same shite.
[33:42] use, and they all give the same shite. It's all the same the same generic
[33:43] It's all the same the same generic reports, the same, oh, we noticed this
[33:45] reports, the same, oh, we noticed this analysis, things that you can find on
[33:47] analysis, things that you can find on any kind of AI slop out there. what you
[33:49] any kind of AI slop out there. what you described to me there, what I heard
[33:50] described to me there, what I heard anyway is there's two points of value
[33:52] anyway is there's two points of value you're getting from your time with that.
[33:53] you're getting from your time with that. I mean, I mean, there's many more, but
[33:55] I mean, I mean, there's many more, but you said you're you're learning and then
[33:57] you said you're you're learning and then you're getting this book edited blog
[33:59] you're getting this book edited blog blog. You're getting a blog edited,
[34:00] blog. You're getting a blog edited, which is the output, and you're getting
[34:02] which is the output, and you're getting learning and you're also really getting
[34:04] learning and you're also really getting connection and all these other things.
[34:05] connection and all these other things. But when I come to when people sort of
[34:07] But when I come to when people sort of think about the value of AI, of course,
[34:08] think about the value of AI, of course, they could use it to learn. But in the
[34:10] they could use it to learn. But in the example I gave of like repeat my revenue
[34:12] example I gave of like repeat my revenue number back to me or do this number, I I
[34:14] number back to me or do this number, I I just care about the output. I could use
[34:15] just care about the output. I could use it to learn. I could say what if the
[34:17] it to learn. I could say what if the revenue number was wrong once you should
[34:19] revenue number was wrong once you should have defined deterministic ways of
[34:21] have defined deterministic ways of knowing those numbers you should not
[34:23] knowing those numbers you should not rely on them even with the terminal
[34:25] rely on them even with the terminal running BQL which I trust I will double
[34:28] running BQL which I trust I will double triple treble check everything just to
[34:30] triple treble check everything just to be sure partly because also the process
[34:32] be sure partly because also the process of learning for me I don't want just a
[34:34] of learning for me I don't want just a report I go like that I want something
[34:37] report I go like that I want something that I fully understand and also
[34:39] that I fully understand and also understand the context around it I don't
[34:41] understand the context around it I don't think that LLM do that and I just don't
[34:44] think that LLM do that and I just don't see them getting
[34:45] see them getting in a way that does that because it's
[34:48] in a way that does that because it's it's just not what they do. And also
[34:50] it's just not what they do. And also there's the other problem of the more
[34:52] there's the other problem of the more detailed the report, the more likely
[34:53] detailed the report, the more likely there are things to be wrong with it. If
[34:55] there are things to be wrong with it. If you are with Matt Hughes, for example, I
[34:57] you are with Matt Hughes, for example, I can trust he's got it right. I can trust
[34:59] can trust he's got it right. I can trust he understood and I can trust that I can
[35:01] he understood and I can trust that I can have a back and forth with him that will
[35:03] have a back and forth with him that will inform me if I've missed something. I
[35:05] inform me if I've missed something. I can read the stuff that he's read and
[35:07] can read the stuff that he's read and actually trust him because there's a big
[35:09] actually trust him because there's a big trust part as well. What is the basis of
[35:12] trust part as well. What is the basis of your trust in Matt? Could it be his
[35:15] your trust in Matt? Could it be his historical performance?
[35:16] historical performance? >> I mean, yes.
[35:17] >> I mean, yes. >> Okay.
[35:18] >> Okay. >> And also the fact we've learned half of
[35:20] >> And also the fact we've learned half of this stuff together,
[35:20] this stuff together, >> but but tenure tenure doesn't
[35:22] >> but but tenure tenure doesn't necessarily There's probably people, you
[35:24] necessarily There's probably people, you know, for 15 years who you also don't
[35:25] know, for 15 years who you also don't trust. Yes.
[35:26] trust. Yes. >> So, I think I was trying to figure out
[35:27] >> So, I think I was trying to figure out like what is the what is the thing
[35:28] like what is the what is the thing that's causing humans to trust another
[35:30] that's causing humans to trust another thing. And I guess it would be continual
[35:32] thing. And I guess it would be continual delivery of a commitment made of sorts.
[35:35] delivery of a commitment made of sorts. And so with Claude for example on simple
[35:38] And so with Claude for example on simple tasks as we've seen from this
[35:39] tasks as we've seen from this hallucination leaderboard it continually
[35:41] hallucination leaderboard it continually delivers for people and that's why we've
[35:44] delivers for people and that's why we've seen the fast
[35:44] seen the fast >> I mean is that what that board says
[35:46] >> I mean is that what that board says >> well it's it's saying like is it getting
[35:48] >> well it's it's saying like is it getting it wrong is it hallucinating
[35:49] it wrong is it hallucinating >> simple task how are those defined
[35:51] >> simple task how are those defined >> I I don't know
[35:52] >> I I don't know >> that's the thing though because this is
[35:53] >> that's the thing though because this is actually a very very illustrative thing
[35:55] actually a very very illustrative thing of the AI industry they are the what
[35:59] of the AI industry they are the what aboutist masters they have like well
[36:01] aboutist masters they have like well look we got this we got this benchmark
[36:03] look we got this we got this benchmark that says we're good at this and look
[36:04] that says we're good at this and look the numbers higher What's the number
[36:05] the numbers higher What's the number mean? No. What does that mean? And I'm
[36:08] mean? No. What does that mean? And I'm not using this as a critic against you.
[36:10] not using this as a critic against you. It's
[36:11] It's >> when you can't give a direct answer, you
[36:13] >> when you can't give a direct answer, you give a side answer. When you as the LLM
[36:15] give a side answer. When you as the LLM industry want to prove your worth, you
[36:17] industry want to prove your worth, you can't just be like just use the product.
[36:19] can't just be like just use the product. When the first iPhone came out, go was
[36:21] When the first iPhone came out, go was Penn State at the time. Oh, I felt like
[36:23] Penn State at the time. Oh, I felt like the uh apes at the beginning of 2001.
[36:25] the uh apes at the beginning of 2001. official voicemail. It was
[36:28] official voicemail. It was immediate. And I showed it to tech
[36:29] immediate. And I showed it to tech friends. I showed it to the most normal
[36:31] friends. I showed it to the most normal people in the world. And everyone was
[36:33] people in the world. And everyone was like, "Holy this is They were on
[36:34] like, "Holy this is They were on razors. They were on Nokia 3210s. It was
[36:37] razors. They were on Nokia 3210s. It was obvious the value." Amazon Web Services,
[36:39] obvious the value." Amazon Web Services, same deal.
[36:40] same deal. >> It wasn't obvious though.
[36:41] >> It wasn't obvious though. >> Yes, it was. I mean, I bought it
[36:42] >> Yes, it was. I mean, I bought it >> to you. To you, it was.
[36:43] >> to you. To you, it was. >> It was. And I also showed it to a bunch
[36:45] >> It was. And I also showed it to a bunch of people because I'm aware that I had
[36:47] of people because I'm aware that I had bias when I just love gadgets.
[36:48] bias when I just love gadgets. >> But but I remember the famous Steve
[36:50] >> But but I remember the famous Steve Balmer who was the CEO of Microsoft
[36:53] Balmer who was the CEO of Microsoft interview where he was told about the
[36:55] interview where he was told about the iPhone and he bursts out laughing.
[36:59] iPhone and he bursts out laughing. [laughter]
[37:00] [laughter] $500 fully subsidized with a plan. I
[37:04] $500 fully subsidized with a plan. I said that is the most expensive phone in
[37:06] said that is the most expensive phone in the world and it doesn't appeal to
[37:08] the world and it doesn't appeal to business customers because it doesn't
[37:10] business customers because it doesn't have a keyboard which makes it not a
[37:12] have a keyboard which makes it not a very good email machine. You can get a
[37:14] very good email machine. You can get a Motorola Q phone now for $99. It's a
[37:18] Motorola Q phone now for $99. It's a very capable machine. It'll do music.
[37:20] very capable machine. It'll do music. It'll do internet. It'll do email. It'll
[37:23] It'll do internet. It'll do email. It'll do instant messaging. So, I I kind of
[37:26] do instant messaging. So, I I kind of look at that and I say, "Well, I like
[37:28] look at that and I say, "Well, I like our strategy. I like it a lot.
[37:31] our strategy. I like it a lot. >> He burst out laughing, mocking it
[37:32] >> He burst out laughing, mocking it because it was so disruptive. It was way
[37:34] because it was so disruptive. It was way more expensive
[37:35] more expensive >> and it was way different. No keyboard.
[37:37] >> and it was way different. No keyboard. >> Well, phones used to be insanely
[37:39] >> Well, phones used to be insanely expensive and the carriers would cover
[37:41] expensive and the carriers would cover them, but you had to sign a long
[37:42] them, but you had to sign a long contract. You were still spending 500
[37:43] contract. You were still spending 500 bucks. But the thing I'm getting at is
[37:45] bucks. But the thing I'm getting at is you didn't have to explain to someone
[37:46] you didn't have to explain to someone why perhaps you'd have to get past the
[37:48] why perhaps you'd have to get past the cost part, but you could just be like,
[37:50] cost part, but you could just be like, "Look how good this is." And then once
[37:51] "Look how good this is." And then once the app was the iPhone 3G with the App
[37:53] the app was the iPhone 3G with the App Store, people were like, "Oh this
[37:55] Store, people were like, "Oh this could actually change things." mobile
[37:56] could actually change things." mobile web. Even though it was a monstrosity,
[37:58] web. Even though it was a monstrosity, it was so bad at first. Even then, you
[38:00] it was so bad at first. Even then, you could get your emails and you could just
[38:02] could get your emails and you could just look at them. Point is, Blackberries
[38:03] look at them. Point is, Blackberries were also expensive and were still
[38:04] were also expensive and were still actually kind of cool, but the way they
[38:06] actually kind of cool, but the way they worked was not like consumer software.
[38:08] worked was not like consumer software. They didn't have the classic GUI.
[38:09] They didn't have the classic GUI. iPhones felt like that. It felt like an
[38:11] iPhones felt like that. It felt like an a cell phone designed even like a
[38:14] a cell phone designed even like a computer. It was obvious. It was obvious
[38:16] computer. It was obvious. It was obvious from the beginning. Everyone I was I was
[38:18] from the beginning. Everyone I was I was dating a girl in the center of
[38:20] dating a girl in the center of Pennsylvania at the time and everyone I
[38:21] Pennsylvania at the time and everyone I showed it to was like, "Wow, this is
[38:22] showed it to was like, "Wow, this is incredible." That to me is the obvious
[38:24] incredible." That to me is the obvious thing with AI to this day when you're
[38:27] thing with AI to this day when you're like, "Okay, why is it so amazing?"
[38:29] like, "Okay, why is it so amazing?" People still dither. People are still
[38:30] People still dither. People are still like, "Yeah, you can't run a business
[38:33] like, "Yeah, you can't run a business fully with it without this weird system
[38:36] fully with it without this weird system of pulleys and levers and such."
[38:38] of pulleys and levers and such." >> But how come then when you look at the
[38:40] >> But how come then when you look at the stats around ChachiBT's growth,
[38:43] stats around ChachiBT's growth, >> 100 million active users in just the
[38:46] >> 100 million active users in just the first 60 days after launching? For
[38:47] first 60 days after launching? For comparison, Tik Tok took 9 months.
[38:49] comparison, Tik Tok took 9 months. Instagram took 2.5 years. And the
[38:51] Instagram took 2.5 years. And the internet itself for the worldwide web
[38:52] internet itself for the worldwide web took roughly 7 years to reach that
[38:54] took roughly 7 years to reach that scale. Over 60% of the US adults are
[38:56] scale. Over 60% of the US adults are integrated into AI tools in their daily
[38:59] integrated into AI tools in their daily and regular routines within 3 years of
[39:01] and regular routines within 3 years of the launch, reaching a 40% of the
[39:03] the launch, reaching a 40% of the population. And that same milestone took
[39:05] population. And that same milestone took the internet 5 years and personal
[39:07] the internet 5 years and personal computers nearly 12.
[39:09] computers nearly 12. >> Okay. So like this is the I think this
[39:10] >> Okay. So like this is the I think this is the part that's giving me dissonance
[39:12] is the part that's giving me dissonance is like when I showed my fiance chachi
[39:14] is like when I showed my fiance chachi okay it was didn't [clears throat]
[39:15] okay it was didn't [clears throat] really work
[39:16] really work >> but as a sole entrepreneur who English
[39:19] >> but as a sole entrepreneur who English isn't her first language
[39:20] isn't her first language >> who has to write lots of text lots of
[39:22] >> who has to write lots of text lots of copy and generate lots of images and was
[39:24] copy and generate lots of images and was paying a graphic designer to help her
[39:25] paying a graphic designer to help her make um certain images that she you know
[39:27] make um certain images that she you know couldn't make herself because she
[39:29] couldn't make herself because she doesn't have the skills.
[39:30] doesn't have the skills. >> She would describe it as being
[39:33] >> She would describe it as being transformative for her business. What
[39:36] transformative for her business. What I'm hearing from you is that it's not
[39:37] I'm hearing from you is that it's not transformative and there's no value in
[39:39] transformative and there's no value in it for people. But she if she was sat
[39:40] it for people. But she if she was sat here transformative,
[39:42] here transformative, would she pay the per million token
[39:44] would she pay the per million token rate? Would she pay the actual rate? Cuz
[39:46] rate? Would she pay the actual rate? Cuz that's the thing. If this was sold at
[39:48] that's the thing. If this was sold at its honest cost. Yeah.
[39:49] its honest cost. Yeah. >> I would actually if and people were
[39:51] >> I would actually if and people were reacting like that and they were paying
[39:52] reacting like that and they were paying 23 $4 every time they did something and
[39:54] 23 $4 every time they did something and they were genuinely happy. That might be
[39:56] they were genuinely happy. That might be an argument.
[39:56] an argument. >> What is the what would be the honest
[39:58] >> What is the what would be the honest cost if they weren't sub
[39:59] cost if they weren't sub >> the actual per million token cost? The
[40:01] >> the actual per million token cost? The actual API cost they should char.
[40:02] actual API cost they should char. >> Do you know how much that is relative to
[40:04] >> Do you know how much that is relative to God? Depends on it depends on the model.
[40:06] God? Depends on it depends on the model. But there's actually kind of a point I
[40:09] But there's actually kind of a point I want to make about the thing you said
[40:11] want to make about the thing you said with the internet earlier. So when I
[40:12] with the internet earlier. So when I first got on the internet 33.4 kilobits
[40:15] first got on the internet 33.4 kilobits a second modem even back then I was like
[40:18] a second modem even back then I was like if this was faster and that was
[40:20] if this was faster and that was like immediate just like if this was
[40:22] like immediate just like if this was faster cuz it was slow. You go on like
[40:24] faster cuz it was slow. You go on like happy puppy or something download take
[40:26] happy puppy or something download take all bloody day waiting for share word to
[40:28] all bloody day waiting for share word to download immediately like if I could do
[40:30] download immediately like if I could do this faster it would be better. And even
[40:31] this faster it would be better. And even back then I'm like, man, you could
[40:32] back then I'm like, man, you could probably do video camera stuff with this
[40:34] probably do video camera stuff with this stuff that eventually happened. And
[40:36] stuff that eventually happened. And actually, there's this guy called Jim
[40:37] actually, there's this guy called Jim Cavell from Goldman Sachs in a report he
[40:39] Cavell from Goldman Sachs in a report he did in 2024 that was geni too much spend
[40:42] did in 2024 that was geni too much spend for not enough return. Paraphrasing
[40:43] for not enough return. Paraphrasing there. And he made the point that in the
[40:45] there. And he made the point that in the run-up to the iPhone, there was
[40:48] run-up to the iPhone, there was thousands of presentations that when GSM
[40:50] thousands of presentations that when GSM radios get smaller, when Bluetooth
[40:51] radios get smaller, when Bluetooth radios get smaller, when Wi-Fi radios
[40:53] radios get smaller, when Wi-Fi radios get smaller, it is inevitable that we
[40:55] get smaller, it is inevitable that we will get something like this. And then
[40:57] will get something like this. And then he said that there is no such path for
[41:00] he said that there is no such path for AI. There was no road map to AI becoming
[41:03] AI. There was no road map to AI becoming this thing that they promised. And I
[41:05] this thing that they promised. And I must be clear, if these companies had
[41:07] must be clear, if these companies had gone out there and are like, "Yeah, this
[41:09] gone out there and are like, "Yeah, this is interesting cloud software. It's
[41:10] is interesting cloud software. It's generative. It's really expensive. We're
[41:12] generative. It's really expensive. We're not sure if we can fully not trust it.
[41:15] not sure if we can fully not trust it. Not in the I'm scared way. I mean, just
[41:17] Not in the I'm scared way. I mean, just like we're not sure that this is going
[41:19] like we're not sure that this is going to be a disruptive world changing thing.
[41:22] to be a disruptive world changing thing. It has potential, but we're going to go
[41:23] It has potential, but we're going to go slow. It's really expensive. This is an
[41:25] slow. It's really expensive. This is an R&D effort. We're not going to expose
[41:27] R&D effort. We're not going to expose consumers to it." and actually being
[41:29] consumers to it." and actually being like called them like I don't know
[41:30] like called them like I don't know language models and no no generative AI
[41:33] language models and no no generative AI stuff just being not even call it
[41:35] stuff just being not even call it because it isn't AI it's not autonomous
[41:37] because it isn't AI it's not autonomous it's not smart I actually might respect
[41:39] it's not smart I actually might respect it but this is not they've gone out
[41:41] it but this is not they've gone out there since 2023 and said it was 2022
[41:43] there since 2023 and said it was 2022 this is the best thing since sliced
[41:45] this is the best thing since sliced bread this is changing everything this
[41:47] bread this is changing everything this is going to do all your work this is
[41:48] is going to do all your work this is going to take your job you're going to
[41:50] going to take your job you're going to talk to Bing and it's going to tell you
[41:51] talk to Bing and it's going to tell you to leave your wife all of these crazy
[41:53] to leave your wife all of these crazy things and what's funny is when the
[41:55] things and what's funny is when the writer uh Kevin Roose I think it was
[41:58] writer uh Kevin Roose I think it was He was speaking to Kevin Scott, the CTO
[42:00] He was speaking to Kevin Scott, the CTO of Microsoft, about it. And Kevin Scott
[42:01] of Microsoft, about it. And Kevin Scott goes, you know, I'm just glad we're
[42:03] goes, you know, I'm just glad we're having this conversation. Instead of
[42:04] having this conversation. Instead of being like, "Settle down, Beas. It's a
[42:07] being like, "Settle down, Beas. It's a website. The website told you something.
[42:09] website. The website told you something. It's just LLM." They talked it up. And
[42:11] It's just LLM." They talked it up. And that's because everyone is talking about
[42:13] that's because everyone is talking about what they wish this was. Rather than
[42:14] what they wish this was. Rather than talking about what it can actually do.
[42:16] talking about what it can actually do. This makes it scary to people
[42:18] This makes it scary to people deliberately. So, it makes it
[42:20] deliberately. So, it makes it environmentally destructive. Look at the
[42:22] environmentally destructive. Look at the gas turbines poisoning black
[42:23] gas turbines poisoning black neighborhoods. I think it's in
[42:24] neighborhoods. I think it's in Louisiana. It's one of Musk's data
[42:26] Louisiana. It's one of Musk's data centers. Look at the incredible energy
[42:28] centers. Look at the incredible energy draws. It is raising power bills and
[42:31] draws. It is raising power bills and also it is creating inflation across all
[42:33] also it is creating inflation across all consumer electronics because of the
[42:35] consumer electronics because of the massive RAM.
[42:36] massive RAM. >> You know what's interesting? I almost
[42:38] >> You know what's interesting? I almost feel like so much of what you're saying
[42:40] feel like so much of what you're saying is true and also it can be true that
[42:45] is true and also it can be true that this technology is going to profoundly
[42:47] this technology is going to profoundly change the world. And I think like you
[42:49] change the world. And I think like you know I think back to the early days of
[42:52] know I think back to the early days of the internet is maybe the closest
[42:53] the internet is maybe the closest analogy we have of you know in the com
[42:55] analogy we have of you know in the com bubble. you know, you wrote this great
[42:56] bubble. you know, you wrote this great essay.
[42:57] essay. >> Yes. Yes.
[42:58] >> Yes. Yes. >> Which I found really funny um especially
[43:00] >> Which I found really funny um especially the name the rot economy and you talked
[43:03] the name the rot economy and you talked about the rotcom bubble.
[43:04] about the rotcom bubble. >> Yes.
[43:05] >> Yes. >> Talking about how AI is of less value
[43:08] >> Talking about how AI is of less value than people think.
[43:10] than people think. >> And in that in the sort of com bubble,
[43:12] >> And in that in the sort of com bubble, what you saw is huge hype, people
[43:14] what you saw is huge hype, people overselling the capabilities of their
[43:16] overselling the capabilities of their websites and what they were building.
[43:18] websites and what they were building. But in the wake of the dotcom bubble,
[43:21] But in the wake of the dotcom bubble, yes, 90% of stuff went to zero,
[43:24] yes, 90% of stuff went to zero, >> but you had generational companies born
[43:26] >> but you had generational companies born that changed the world,
[43:27] that changed the world, >> right?
[43:28] >> right? >> And so I I do I kind of and that's what
[43:31] >> And so I I do I kind of and that's what bubbles do, right? Huge hype,
[43:33] bubbles do, right? Huge hype, overinvestment, investors get crazy,
[43:35] overinvestment, investors get crazy, delusional. They think it's everything's
[43:36] delusional. They think it's everything's going to change. At the same time, you
[43:39] going to change. At the same time, you do have skeptics
[43:41] do have skeptics >> in these moments. The the dot bubble had
[43:43] >> in these moments. The the dot bubble had I mean the internet itself had the
[43:44] I mean the internet itself had the biggest skeptics in 1998. Nobel Prize
[43:47] biggest skeptics in 1998. Nobel Prize winning economist Paul Krugman said by
[43:50] winning economist Paul Krugman said by 2005 or so it will become clear that the
[43:53] 2005 or so it will become clear that the internet's impact on the economy has
[43:54] internet's impact on the economy has been no greater than the fax machine. In
[43:56] been no greater than the fax machine. In 1995 astrophysicist Clifford stool
[43:59] 1995 astrophysicist Clifford stool famously I wrote about this in my book
[44:02] famously I wrote about this in my book wrote famously in Newsweek. Do our
[44:04] wrote famously in Newsweek. Do our computer pundits lack all common sense?
[44:06] computer pundits lack all common sense? The truth is no online database will
[44:09] The truth is no online database will replace your daily newspaper. No CDROM
[44:11] replace your daily newspaper. No CDROM can take the place of a competent
[44:13] can take the place of a competent teacher. Commerce and businesses will
[44:15] teacher. Commerce and businesses will shift from offices and malls to networks
[44:17] shift from offices and malls to networks and modems. Bologoney. So, how come my
[44:19] and modems. Bologoney. So, how come my local mall does a roaring business and
[44:22] local mall does a roaring business and the cyber mall gets zero business? And
[44:25] the cyber mall gets zero business? And then I'll give you one more from
[44:26] then I'll give you one more from Krueger, who was the award-winning
[44:28] Krueger, who was the award-winning economist. He said, "The growth of the
[44:30] economist. He said, "The growth of the internet will slow drastically as it
[44:32] internet will slow drastically as it becomes apparent most people have
[44:34] becomes apparent most people have nothing to say to each other."
[44:36] nothing to say to each other." That's that that that may actually be
[44:38] That's that that that may actually be the worst one of those predict like hang
[44:41] the worst one of those predict like hang around any bar in middle America.
[44:43] around any bar in middle America. Honestly, the best conversation,
[44:44] Honestly, the best conversation, >> but it's just all the same thing.
[44:46] >> but it's just all the same thing. >> I actually So, Clifford Stall actually
[44:48] >> I actually So, Clifford Stall actually his piece was interesting cuz that there
[44:49] his piece was interesting cuz that there were some boner points in it, but he
[44:51] were some boner points in it, but he made points about how like an
[44:52] made points about how like an overwhelming amount of bad information
[44:54] overwhelming amount of bad information out there is bad for society. He's
[44:55] out there is bad for society. He's completely right saying how online
[44:56] completely right saying how online education would not be a great
[44:59] education would not be a great replacement for regular education. I
[45:01] replacement for regular education. I think we've seen that. But there is an
[45:03] think we've seen that. But there is an economic difference that's vastly it's
[45:05] economic difference that's vastly it's just completely different. So.com bubble
[45:07] just completely different. So.com bubble was actually two bubbles. There was the
[45:09] was actually two bubbles. There was the website bubble which was just trash on
[45:11] website bubble which was just trash on trash on trash. It was just like I think
[45:13] trash on trash. It was just like I think what was it? Excite at home bought a
[45:16] what was it? Excite at home bought a eury incard company for like a billion
[45:18] eury incard company for like a billion dollars. It was insane crap happening
[45:20] dollars. It was insane crap happening that was so small. The big thing that
[45:23] that was so small. The big thing that people are thinking about is the dark
[45:24] people are thinking about is the dark fiber.
[45:25] fiber. >> Dark fiber. dark fiber was all of the
[45:27] >> Dark fiber. dark fiber was all of the wires that put in the ground thinking
[45:29] wires that put in the ground thinking we're going to have all this demand for
[45:30] we're going to have all this demand for internet and it turned out that demand
[45:32] internet and it turned out that demand for internet I think the analyst
[45:35] for internet I think the analyst estimate was it was doubling every 90
[45:37] estimate was it was doubling every 90 days when it was doing that every 6 to
[45:39] days when it was doing that every 6 to 12 months maybe maybe longer and just
[45:42] 12 months maybe maybe longer and just thus there was a massive overbuild of
[45:43] thus there was a massive overbuild of fiber optic cable and indeed the
[45:46] fiber optic cable and indeed the transmission stations and such just
[45:48] transmission stations and such just simplifying to bring that to people's
[45:50] simplifying to bring that to people's houses and there was the assumption that
[45:52] houses and there was the assumption that well that would all get lit up and
[45:53] well that would all get lit up and people would want it immediately didn't
[45:54] people would want it immediately didn't really
[45:55] really Now the post.com bubble thing people say
[45:57] Now the post.com bubble thing people say is well but after that there was demand
[45:59] is well but after that there was demand from the internet. That's the thing
[46:01] from the internet. That's the thing though that's very different to demand
[46:04] though that's very different to demand for generative AI. Right now the demand
[46:06] for generative AI. Right now the demand we have for generative AI is
[46:08] we have for generative AI is predominantly subsidized. Just let's
[46:10] predominantly subsidized. Just let's start there.
[46:10] start there. >> Yeah
[46:11] >> Yeah >> predominantly subsidized and most people
[46:12] >> predominantly subsidized and most people experience it are not paying the real
[46:14] experience it are not paying the real cost.
[46:15] cost. >> I agree.
[46:15] >> I agree. >> On top of that we already have all of
[46:18] >> On top of that we already have all of the possible marketing in the world. We
[46:21] the possible marketing in the world. We have the largest, most disingenuous
[46:23] have the largest, most disingenuous marketing campaign in the history of
[46:25] marketing campaign in the history of man, pushing this up the hill. We have
[46:27] man, pushing this up the hill. We have the apex predator of cloud software,
[46:30] the apex predator of cloud software, Microsoft. They can only get singledigit
[46:33] Microsoft. They can only get singledigit billions from selling AI software. And
[46:36] billions from selling AI software. And Christ almighty, outside of OpenAI and
[46:38] Christ almighty, outside of OpenAI and Anthropic, we barely get $22 billion.
[46:40] Anthropic, we barely get $22 billion. And the thing is, $22 billion is a large
[46:42] And the thing is, $22 billion is a large amount to you and me. It's not a large
[46:44] amount to you and me. It's not a large amount of money when you spent a
[46:46] amount of money when you spent a trillion plus dollars. When you have
[46:47] trillion plus dollars. When you have anthropic and open AI with $1.1 trillion
[46:50] anthropic and open AI with $1.1 trillion worth of cloud commitments and on top of
[46:52] worth of cloud commitments and on top of that, how does this turn into a post.com
[46:55] that, how does this turn into a post.com bubble thing? A data center built today
[46:57] bubble thing? A data center built today is going to be as expensive to run in
[46:59] is going to be as expensive to run in 2050 as it is today unless there's some
[47:01] 2050 as it is today unless there's some breakthrough in electricity. But again,
[47:04] breakthrough in electricity. But again, that's not happening with AI. AI is not
[47:06] that's not happening with AI. AI is not doing that unless there's some
[47:08] doing that unless there's some breakthrough in GPU technology. But we
[47:10] breakthrough in GPU technology. But we already have Broadcom, Nvidia, etched.
[47:13] already have Broadcom, Nvidia, etched. We have every major chip company ARM
[47:15] We have every major chip company ARM trying to do something about this. And
[47:17] trying to do something about this. And no one seems to magically be able to
[47:19] no one seems to magically be able to make this profitable or indeed even less
[47:22] make this profitable or indeed even less costly. Even Nvidia with Vera Rubin,
[47:25] costly. Even Nvidia with Vera Rubin, their more expensive new GPU system.
[47:27] their more expensive new GPU system. Even then, they're like, "Yeah, 10x more
[47:29] Even then, they're like, "Yeah, 10x more efficient. It's uh more dollars per
[47:30] efficient. It's uh more dollars per megawatt." They're all koi about it.
[47:32] megawatt." They're all koi about it. They don't just say, "Yeah, we worked
[47:34] They don't just say, "Yeah, we worked with OpenAI and Anthropic and we found
[47:35] with OpenAI and Anthropic and we found it reduced our cost by 50%." Easiest
[47:37] it reduced our cost by 50%." Easiest thing in the world if it was true. And
[47:39] thing in the world if it was true. And that's because it's not happening. And
[47:41] that's because it's not happening. And this isn't a case where
[47:43] this isn't a case where >> So are you saying there's not going to
[47:44] >> So are you saying there's not going to be the demand for let's say let's you
[47:47] be the demand for let's say let's you know there's different types of AI
[47:48] know there's different types of AI generative AI we
[47:50] generative AI we >> Yeah. And actually that's a good point
[47:51] >> Yeah. And actually that's a good point to make. The reason they use the term
[47:53] to make. The reason they use the term artificial intelligence is so everyone
[47:54] artificial intelligence is so everyone would lump everything into it.
[47:56] would lump everything into it. >> They [clears throat] would lump uh
[47:57] >> They [clears throat] would lump uh protein folding nothing to do with LLMs.
[47:59] protein folding nothing to do with LLMs. Robotics not LLM.
[48:01] Robotics not LLM. >> Autonomous weapons even horrible as they
[48:03] >> Autonomous weapons even horrible as they are not LLMs because you couldn't trust
[48:05] are not LLMs because you couldn't trust them. But they've mushed everything into
[48:07] them. But they've mushed everything into AI so that when you say, "Well, AI
[48:10] AI so that when you say, "Well, AI can't," they'll go, "Um, um, sir, you
[48:12] can't," they'll go, "Um, um, sir, you forgot to give us homework and also AI
[48:14] forgot to give us homework and also AI it's working on curing cancer." When
[48:16] it's working on curing cancer." When it's just like, "No, that's not LLM.
[48:18] it's just like, "No, that's not LLM. Stop giving them credit."
[48:19] Stop giving them credit." >> The similarity though is they all need
[48:20] >> The similarity though is they all need GPUs, all these.
[48:22] GPUs, all these. >> And that's the funny thing. All those
[48:23] >> And that's the funny thing. All those data centers that we're building, all of
[48:25] data centers that we're building, all of them are for just generative AI. They're
[48:28] them are for just generative AI. They're not for all of the other stuff. They're
[48:31] not for all of the other stuff. They're not for the cool AI has been
[48:33] not for the cool AI has been around for a long time. Google. A lot of
[48:35] around for a long time. Google. A lot of the good stuff that comes out of Google
[48:37] the good stuff that comes out of Google from the search side is AI but
[48:39] from the search side is AI but pre-generative.
[48:40] pre-generative. >> How would you run the the type of AI
[48:43] >> How would you run the the type of AI that sits in a robot? Let's say one of
[48:45] that sits in a robot? Let's say one of the Optimus robots if you didn't have a
[48:46] the Optimus robots if you didn't have a GPU.
[48:47] GPU. >> So Matic Matic has this cleaning robot
[48:49] >> So Matic Matic has this cleaning robot for example. That thing is not got a
[48:52] for example. That thing is not got a little GPU in it. What it has and may
[48:54] little GPU in it. What it has and may indeed have used some GPUs but no year
[48:57] indeed have used some GPUs but no year as many as they need for generative AI
[49:00] as many as they need for generative AI to run the data feed training data into
[49:02] to run the data feed training data into it so it's able to clean a house. But
[49:03] it so it's able to clean a house. But when the little buggers going around
[49:04] when the little buggers going around cleaning my floor, turdsly I call him,
[49:06] cleaning my floor, turdsly I call him, it goes around mopping my floor, it's
[49:08] it goes around mopping my floor, it's not like burning money the whole time.
[49:10] not like burning money the whole time. But when it comes to these massive
[49:12] But when it comes to these massive amount of data center, sighteline
[49:14] amount of data center, sighteline climate said in February there's 190
[49:16] climate said in February there's 190 gawatts of data centers under in
[49:18] gawatts of data centers under in planning. Don't know about under
[49:19] planning. Don't know about under construction that works out if about 12
[49:22] construction that works out if about 12 million megawatt that's what like $1.6
[49:24] million megawatt that's what like $1.6 trillion to3 trillion a year in annual
[49:26] trillion to3 trillion a year in annual demand you'd need for that. We don't
[49:28] demand you'd need for that. We don't even have $130 billion worth of annual
[49:30] even have $130 billion worth of annual demand. And people say, well, it will
[49:32] demand. And people say, well, it will grow. how when most of the demand is
[49:34] grow. how when most of the demand is coming from Amazon feeding money to open
[49:37] coming from Amazon feeding money to open AAI or anthropic, Microsoft feeding
[49:39] AAI or anthropic, Microsoft feeding money to OpenAI and Anthropic, Google
[49:41] money to OpenAI and Anthropic, Google feeding money to Open AI and anthrop
[49:42] feeding money to Open AI and anthrop well hasn't fed it to Open AI yet, but
[49:45] well hasn't fed it to Open AI yet, but they're a pretty big customer, billions
[49:46] they're a pretty big customer, billions of dollars. The conside is that we are
[49:49] of dollars. The conside is that we are building these effiges to capitalism,
[49:51] building these effiges to capitalism, these giant GPU data centers, and people
[49:54] these giant GPU data centers, and people are being told, well, it's for AI, you
[49:57] are being told, well, it's for AI, you know, the thing that's done all this
[49:58] know, the thing that's done all this other stuff that's unrelated. Or the
[50:00] other stuff that's unrelated. Or the worst thing I've seen is like, oh, you
[50:01] worst thing I've seen is like, oh, you don't like you like online banking.
[50:03] don't like you like online banking. Well, you do like data centers. There's
[50:04] Well, you do like data centers. There's a big difference between a data center
[50:06] a big difference between a data center for regular nonGPU compute for standing
[50:08] for regular nonGPU compute for standing up a server, a content delivery system
[50:10] up a server, a content delivery system like Akami or something that brings the
[50:12] like Akami or something that brings the website to you or how Meta runs
[50:14] website to you or how Meta runs Facebook. That is not the same. It takes
[50:17] Facebook. That is not the same. It takes way less power, mostly CPUdriven
[50:19] way less power, mostly CPUdriven compared to these giant GPU data centers
[50:21] compared to these giant GPU data centers that offer one thing, one thing only.
[50:24] that offer one thing, one thing only. >> But I was doing the the research and
[50:26] >> But I was doing the the research and looking at some of these notes here. It
[50:27] looking at some of these notes here. It does say that for tougher types of AI
[50:29] does say that for tougher types of AI systems designed to solve concrete
[50:31] systems designed to solve concrete physics, biology, and spatial problems,
[50:32] physics, biology, and spatial problems, they require some of the most intense
[50:34] they require some of the most intense data center infrastructure on the
[50:36] data center infrastructure on the planet.
[50:37] planet. >> Yeah.
[50:37] >> Yeah. >> AI systems like Deep Mind's AlphaFold,
[50:39] >> AI systems like Deep Mind's AlphaFold, the protein folding company
[50:41] the protein folding company >> used for genomic sequencing and climate
[50:44] >> used for genomic sequencing and climate forecasting, etc. run on high
[50:46] forecasting, etc. run on high performance computing clusters. These
[50:47] performance computing clusters. These require immense precision and continuous
[50:49] require immense precision and continuous heavy computing data centers.
[50:51] heavy computing data centers. >> Yeah. Training the brains for
[50:53] >> Yeah. Training the brains for self-driving cars requires billions of
[50:55] self-driving cars requires billions of miles of simulated physics environments.
[50:59] miles of simulated physics environments. The AI isn't generating text. It's
[51:01] The AI isn't generating text. It's learning to navigate 3D spaces and
[51:03] learning to navigate 3D spaces and gravity and relies on data centers,
[51:05] gravity and relies on data centers, >> right? And the thing is those data
[51:07] >> right? And the thing is those data centers, they might have GPUs in them.
[51:09] centers, they might have GPUs in them. We had GPUs used for this HPC, the high
[51:12] We had GPUs used for this HPC, the high performance computing before generative
[51:14] performance computing before generative AI. And yeah, that's how AI has been
[51:16] AI. And yeah, that's how AI has been trained before. That's how Tesla did.
[51:18] trained before. That's how Tesla did. believe they've had their own data
[51:20] believe they've had their own data centers when it comes to training the
[51:21] centers when it comes to training the autopilot system for better or for
[51:23] autopilot system for better or for worse. That's how we've done it before.
[51:25] worse. That's how we've done it before. Again, that is not why we're building
[51:27] Again, that is not why we're building these data centers. These data centers
[51:28] these data centers. These data centers are being built to sell to AI generative
[51:31] are being built to sell to AI generative AI companies to either train systems or
[51:34] AI companies to either train systems or run inference. These things are being
[51:36] run inference. These things are being built in this brainless way where it's
[51:39] built in this brainless way where it's just well actually maybe this is a good
[51:42] just well actually maybe this is a good way of illustrating the con because
[51:44] way of illustrating the con because everyone saw Google, Microsoft, Amazon
[51:48] everyone saw Google, Microsoft, Amazon and Meta give Nvidia over call it 800
[51:52] and Meta give Nvidia over call it 800 something billion dollars
[51:54] something billion dollars because everyone saw that they went well
[51:56] because everyone saw that they went well they wouldn't do that for no reason.
[51:57] they wouldn't do that for no reason. They went we got to build more of these
[51:59] They went we got to build more of these things. There must be all this demand.
[52:00] things. There must be all this demand. Even though the demand 70% or more of
[52:04] Even though the demand 70% or more of all that demand comes from these two
[52:06] all that demand comes from these two companies who were funded by these three
[52:08] companies who were funded by these three companies and that's the funny thing.
[52:10] companies and that's the funny thing. The reason that they don't want to break
[52:12] The reason that they don't want to break out their AI revenues is because it will
[52:14] out their AI revenues is because it will become alarmingly obvious that this was
[52:17] become alarmingly obvious that this was the case. It turns out that the only
[52:19] the case. It turns out that the only real big customers cuz it's not like
[52:21] real big customers cuz it's not like they're building a few data centers.
[52:23] they're building a few data centers. They're building trillion plus revenue
[52:26] They're building trillion plus revenue potential. They believe they'll get
[52:28] potential. They believe they'll get speculative. It's entirely speculative.
[52:30] speculative. It's entirely speculative. They're building it because they saw the
[52:31] They're building it because they saw the biggest companies in the world buy a
[52:33] biggest companies in the world buy a bunch of GPUs and they said, "I want in
[52:35] bunch of GPUs and they said, "I want in on that." They must have diverse
[52:36] on that." They must have diverse customers, right? They wouldn't just
[52:38] customers, right? They wouldn't just have two unprofitable fail sons that
[52:40] have two unprofitable fail sons that they're propping up with. Christ,
[52:42] they're propping up with. Christ, they've raised $217 billion just in
[52:45] they've raised $217 billion just in 2026.
[52:47] 2026. >> So, we know that some of the biggest
[52:49] >> So, we know that some of the biggest companies in the world are using AI,
[52:51] companies in the world are using AI, generative AI to write a lot of their
[52:53] generative AI to write a lot of their code.
[52:53] code. >> Mhm.
[52:54] >> Mhm. >> That is a great productivity gain for
[52:56] >> That is a great productivity gain for those companies, right? I mean, have you
[52:58] those companies, right? I mean, have you used Google or Facebook or Instagram or
[53:01] used Google or Facebook or Instagram or GitHub recently because they are
[53:03] GitHub recently because they are catastrophically worse? Amazon Web
[53:05] catastrophically worse? Amazon Web Services went down multiple times
[53:06] Services went down multiple times because of their AI coding tool. How
[53:08] because of their AI coding tool. How >> how is how is Google worse?
[53:10] >> how is how is Google worse? >> Well, I'll tell the story of a real
[53:12] >> Well, I'll tell the story of a real guy called Preaggo Ragavan.
[53:14] guy called Preaggo Ragavan. Previously, one of the heads of ads at
[53:15] Previously, one of the heads of ads at Google in 2019, Google called something
[53:18] Google in 2019, Google called something called a code yellow, which is when they
[53:20] called a code yellow, which is when they said, "We've got a problem." And it was
[53:21] said, "We've got a problem." And it was material weakness in query numbers which
[53:25] material weakness in query numbers which means the amount of times that people
[53:26] means the amount of times that people were searching on Google search. Guy
[53:28] were searching on Google search. Guy called Ben Gomes internal at Google then
[53:30] called Ben Gomes internal at Google then the head of Google search says wait a
[53:32] the head of Google search says wait a minute to increase this number of using
[53:34] minute to increase this number of using Google more.
[53:35] Google more. >> Mhm. We're going to have to I mean you
[53:37] >> Mhm. We're going to have to I mean you what you're suggesting would mean we
[53:39] what you're suggesting would mean we give worse answers because if someone
[53:40] give worse answers because if someone got the answer quickly that would reduce
[53:42] got the answer quickly that would reduce the amount of queries right and people
[53:44] the amount of queries right and people at Google Shashi Tako was another
[53:46] at Google Shashi Tako was another engineer was saying yeah can we please
[53:48] engineer was saying yeah can we please tell Sunda this because this doesn't
[53:50] tell Sunda this because this doesn't seem good. We can't just increase the
[53:52] seem good. We can't just increase the amount of queries. That would just mean
[53:54] amount of queries. That would just mean that people would have to search more
[53:55] that people would have to search more which would make the product worse.
[53:56] which would make the product worse. >> But but it would make them more money.
[53:58] >> But but it would make them more money. You saying you'd show them more ads. So
[54:01] You saying you'd show them more ads. So if you're spending more time on Google
[54:02] if you're spending more time on Google because Google's work,
[54:03] because Google's work, >> but is this linked to AI doing code?
[54:05] >> but is this linked to AI doing code? >> Oh, I'll get there. So
[54:07] >> Oh, I'll get there. So >> this is the problem is is that this guy
[54:10] >> this is the problem is is that this guy called Pragar Ragavan who's the head of
[54:12] called Pragar Ragavan who's the head of ads at the time was pushing pushing and
[54:13] ads at the time was pushing pushing and saying, "No, we need to make more
[54:15] saying, "No, we need to make more queries happen. Got to make it happen."
[54:16] queries happen. Got to make it happen." and Nick Fox who was there as well I
[54:18] and Nick Fox who was there as well I believe was actually taking over Google
[54:19] believe was actually taking over Google search got to make them go up this is
[54:20] search got to make them go up this is our new reality sometime in early 2020
[54:24] our new reality sometime in early 2020 propagar ragavan takes over Google
[54:26] propagar ragavan takes over Google search from then and this is this is
[54:28] search from then and this is this is what I believe can't prove it if you go
[54:30] what I believe can't prove it if you go and look around the various SEO sites
[54:32] and look around the various SEO sites such journal and the various forums
[54:34] such journal and the various forums Google stripped back a lot of the
[54:37] Google stripped back a lot of the suppression of spammy sites so that
[54:39] suppression of spammy sites so that people would be on Google more and then
[54:41] people would be on Google more and then over the course of time Google wanted to
[54:44] over the course of time Google wanted to create more queries and Google search
[54:46] create more queries and Google search became much worse. It's why people
[54:47] became much worse. It's why people always do like plus Reddit or from
[54:49] always do like plus Reddit or from Reddit or what have you. It's because
[54:51] Reddit or what have you. It's because the actual underlying search results of
[54:52] the actual underlying search results of Google had got worse. And then
[54:54] Google had got worse. And then Generative AI came along and Praagar,
[54:56] Generative AI came along and Praagar, wouldn't you know, it gets put to run
[54:58] wouldn't you know, it gets put to run part of Gemini. And Google also was
[55:00] part of Gemini. And Google also was having trouble getting people back on
[55:02] having trouble getting people back on Google. And what did they think they'd
[55:04] Google. And what did they think they'd do? Well, everyone's talking about
[55:05] do? Well, everyone's talking about this AI thing. We'll just put it right
[55:08] this AI thing. We'll just put it right at the top so people have to stay at
[55:09] at the top so people have to stay at Google. And actually, they'll use it
[55:11] Google. And actually, they'll use it more because instead of searching
[55:13] more because instead of searching websites and doing that annoying thing
[55:14] websites and doing that annoying thing where they click away from Google,
[55:16] where they click away from Google, they'll just only use Google. Instead of
[55:18] they'll just only use Google. Instead of generating answers, by which I mean
[55:20] generating answers, by which I mean giving you search results you click
[55:21] giving you search results you click through, now Google is the answer. Is it
[55:24] through, now Google is the answer. Is it right? God know. It might tell you to
[55:25] right? God know. It might tell you to eat rocks, might eat poisonous
[55:27] eat rocks, might eat poisonous mushrooms. Maybe it'll give you a little
[55:29] mushrooms. Maybe it'll give you a little few links you could click through. But
[55:30] few links you could click through. But the ideal situation was that AI was the
[55:33] the ideal situation was that AI was the ultimate form of Google's evil which was
[55:36] ultimate form of Google's evil which was >> But I'm saying here I'm saying here but
[55:37] >> But I'm saying here I'm saying here but that's not the fact that coders could
[55:39] that's not the fact that coders could code on Google that's made Google worse.
[55:41] code on Google that's made Google worse. That's human decisions have made it
[55:42] That's human decisions have made it worse.
[55:43] worse. >> Yes. And then there's the instability of
[55:44] >> Yes. And then there's the instability of Google's platform which is actually I
[55:46] Google's platform which is actually I should have probably led with that a
[55:48] should have probably led with that a problem across the whole tech industry.
[55:50] problem across the whole tech industry. >> Okay. So you're saying that you're
[55:51] >> Okay. So you're saying that you're saying Google is going down more.
[55:53] saying Google is going down more. >> Yes. Google is less stable. Google Docs
[55:55] >> Yes. Google is less stable. Google Docs is a bugfest right now and has been for
[55:57] is a bugfest right now and has been for a while. Google Sheets, same deal. And
[55:59] a while. Google Sheets, same deal. And the thing is, you're right, I'm being a
[56:01] the thing is, you're right, I'm being a little unfair. This is everyone. It's
[56:03] little unfair. This is everyone. It's the same with Microsoft. It's the same
[56:04] the same with Microsoft. It's the same with Amazon. It's the same across.
[56:06] with Amazon. It's the same across. >> How do we quantify that outside of
[56:07] >> How do we quantify that outside of anecdotes? Like, is there a way to
[56:09] anecdotes? Like, is there a way to >> You're right. I mean, GitHub downtime is
[56:11] >> You're right. I mean, GitHub downtime is the best example. Amazon Web Services
[56:13] the best example. Amazon Web Services went down two or three times this year
[56:16] went down two or three times this year because of AI tools. And honestly,
[56:19] because of AI tools. And honestly, you're right. It is kind of hard to
[56:20] you're right. It is kind of hard to quantify outside of anecdotes. But I
[56:22] quantify outside of anecdotes. But I challenge anyone listening to this. Go
[56:23] challenge anyone listening to this. Go and use a website these days and tell me
[56:25] and use a website these days and tell me how well it works. Tell me how buggy it
[56:27] how well it works. Tell me how buggy it is. Tell me how many problems even with
[56:29] is. Tell me how many problems even with my iPhone. The supposed best UX in town.
[56:33] my iPhone. The supposed best UX in town. Even the iPhone is a flipping mess these
[56:35] Even the iPhone is a flipping mess these days.
[56:36] days. >> Okay, so the research says the short
[56:39] >> Okay, so the research says the short answer is yes. Tech downtime and
[56:42] answer is yes. Tech downtime and software outages have demonstrabably
[56:43] software outages have demonstrabably increased over the last few years and
[56:45] increased over the last few years and industry data points directly to the
[56:47] industry data points directly to the explosion of AI assisted coding as a
[56:49] explosion of AI assisted coding as a primary culprit. The problem is hitting
[56:51] primary culprit. The problem is hitting the tech industry from two entirely
[56:52] the tech industry from two entirely different directions. The code itself is
[56:55] different directions. The code itself is getting buggier and the sheer volume of
[56:57] getting buggier and the sheer volume of AI activity is literally crashing the
[56:59] AI activity is literally crashing the underlying infrastructure. Interesting.
[57:01] underlying infrastructure. Interesting. >> Yeah, that's because GitHub people are
[57:03] >> Yeah, that's because GitHub people are just writing a bunch of code, pushing
[57:05] just writing a bunch of code, pushing it, and thus there's just more code on
[57:08] it, and thus there's just more code on there.
[57:08] there. >> That's interesting.
[57:09] >> That's interesting. >> Yeah, it's it's a real mess as well
[57:11] >> Yeah, it's it's a real mess as well because
[57:13] because open source has had this problem as well
[57:14] open source has had this problem as well because it's well-meaning people.
[57:16] because it's well-meaning people. They're like, I learned a bit of code
[57:17] They're like, I learned a bit of code with an LLM. I'm going to go out and do
[57:18] with an LLM. I'm going to go out and do some stuff. I'm going to make this
[57:20] some stuff. I'm going to make this project better. And these people barely
[57:22] project better. And these people barely understand what they're shipping. Or
[57:23] understand what they're shipping. Or maybe they understand a bit of code and
[57:25] maybe they understand a bit of code and they say, "Oh, Dunning Krueger, this
[57:27] they say, "Oh, Dunning Krueger, this I'm going to I'm just
[57:28] I'm going to I'm just like, I can understand some of this."
[57:29] like, I can understand some of this." And now the code's all written and just
[57:31] And now the code's all written and just push it right now. So GitHub is flooded
[57:33] push it right now. So GitHub is flooded with AI code.
[57:34] with AI code. >> This sounds like it's making humans
[57:36] >> This sounds like it's making humans complacent.
[57:37] complacent. >> It is
[57:38] >> It is >> because we're going, "Okay, look, I let
[57:40] >> because we're going, "Okay, look, I let it write the the code for the last 100
[57:42] it write the the code for the last 100 lines and it was broadly right. So the
[57:44] lines and it was broadly right. So the next 100 lines, I won't check them as
[57:46] next 100 lines, I won't check them as much."
[57:46] much." >> Yeah. Yeah. And that's human nature is
[57:48] >> Yeah. Yeah. And that's human nature is to get sort of to take shortcuts to
[57:51] to get sort of to take shortcuts to spend less energy on an activity if you
[57:53] spend less energy on an activity if you can right but the AI's still making the
[57:55] can right but the AI's still making the mistake and we're still making all the
[57:57] mistake and we're still making all the promises of AI that's the thing this
[57:59] promises of AI that's the thing this thing is meant to be this autonomous per
[58:02] thing is meant to be this autonomous per you say it can't be perfect I don't know
[58:04] you say it can't be perfect I don't know based on what Samman has been saying for
[58:06] based on what Samman has been saying for the last few years clammy Sammy has been
[58:08] the last few years clammy Sammy has been promising the world saying this will
[58:09] promising the world saying this will replace software engineers Dario
[58:11] replace software engineers Dario Ammedday Wario himself has been saying
[58:14] Ammedday Wario himself has been saying oh yeah 50% of white collar labor is
[58:17] oh yeah 50% of white collar labor is going to go away in the next few years.
[58:18] going to go away in the next few years. These people are promising the world.
[58:20] These people are promising the world. Again, if they were saying it would be
[58:22] Again, if they were saying it would be smaller and they were like, yeah, it
[58:24] smaller and they were like, yeah, it does have issues and we must be none of
[58:27] does have issues and we must be none of this, oh, what if it wakes up and it's
[58:28] this, oh, what if it wakes up and it's super powerful. Just like, yeah, it's
[58:31] super powerful. Just like, yeah, it's probabilistic. It's going to make
[58:32] probabilistic. It's going to make mistakes and if you don't know what
[58:34] mistakes and if you don't know what you're doing, you don't really know what
[58:35] you're doing, you don't really know what you're looking at, you're going to miss
[58:36] you're looking at, you're going to miss those mistakes and it's going to get
[58:37] those mistakes and it's going to get multiplicatively worse as you go when
[58:40] multiplicatively worse as you go when you don't know what you're doing. So
[58:42] you don't know what you're doing. So yeah, human nature is part of it, but so
[58:45] yeah, human nature is part of it, but so is the marketing. So are the promises.
[58:47] is the marketing. So are the promises. One of the smartest things a business
[58:49] One of the smartest things a business can do is build like a bigger company
[58:52] can do is build like a bigger company without actually hiring like one. But
[58:55] without actually hiring like one. But the problem we all face is that most
[58:56] the problem we all face is that most companies don't have every skill in
[58:58] companies don't have every skill in house. So when I look at the businesses
[59:00] house. So when I look at the businesses seeing real success today, the
[59:02] seeing real success today, the consistent pattern with all of them is
[59:04] consistent pattern with all of them is how quickly they move. They bring in
[59:05] how quickly they move. They bring in specialists with skills in emerging
[59:07] specialists with skills in emerging areas to keep themselves ahead. Even in
[59:09] areas to keep themselves ahead. Even in our company, we spent the last year
[59:10] our company, we spent the last year pulling in talent across areas like AI
[59:13] pulling in talent across areas like AI native strategy, no code builds, and
[59:15] native strategy, no code builds, and product workflows. And we find this
[59:17] product workflows. And we find this talent through our longtime partner,
[59:18] talent through our longtime partner, Fiverr Pro. Their premium service only
[59:21] Fiverr Pro. Their premium service only shows you vetted talent. So, you've
[59:23] shows you vetted talent. So, you've always got the safeguard that anyone you
[59:25] always got the safeguard that anyone you pull in to help you with a complex
[59:27] pull in to help you with a complex project has the skills that you're after
[59:30] project has the skills that you're after and will deliver to the same high
[59:31] and will deliver to the same high standards as your internal team. And
[59:33] standards as your internal team. And most importantly, they'll keep up with
[59:35] most importantly, they'll keep up with the pace. It's a simple strategy, but it
[59:36] the pace. It's a simple strategy, but it lets us stay agile without compromising
[59:38] lets us stay agile without compromising on quality. So, if you need these kind
[59:40] on quality. So, if you need these kind of skills in your business, head to
[59:41] of skills in your business, head to pro.fr.com to find pioneering talent to
[59:44] pro.fr.com to find pioneering talent to fill your business's gaps. That's
[59:46] fill your business's gaps. That's pro.fr.com.
[59:48] pro.fr.com. You know, the little traditional SIM
[59:50] You know, the little traditional SIM card that goes inside of our phones.
[59:51] card that goes inside of our phones. They haven't changed at all since they
[59:54] They haven't changed at all since they were invented in the '90s. You have this
[59:56] were invented in the '90s. You have this physical piece of plastic that means
[59:58] physical piece of plastic that means you're locked into one carrier, one
[01:00:00] you're locked into one carrier, one network, and the second you cross a
[01:00:02] network, and the second you cross a border, that carrier can start charging
[01:00:04] border, that carrier can start charging you whatever they want. But there are
[01:00:06] you whatever they want. But there are alternatives, and today's sponsor, SY,
[01:00:08] alternatives, and today's sponsor, SY, is one of them. It's an eim app that
[01:00:11] is one of them. It's an eim app that gives you a safe and secure data
[01:00:12] gives you a safe and secure data connection in over 200 destinations. All
[01:00:16] connection in over 200 destinations. All of their eims have built-in cyber
[01:00:17] of their eims have built-in cyber security, which is great if you're
[01:00:19] security, which is great if you're traveling for work and looking at
[01:00:20] traveling for work and looking at confidential material. I've been using
[01:00:22] confidential material. I've been using SY whenever I travel because the
[01:00:24] SY whenever I travel because the connection is always reliable and it
[01:00:26] connection is always reliable and it saves me a ton of roaming fees. It also
[01:00:28] saves me a ton of roaming fees. It also means I don't have to deal with all of
[01:00:30] means I don't have to deal with all of the faf that surrounds sorting out a SIM
[01:00:32] the faf that surrounds sorting out a SIM everywhere I go. If you want to give it
[01:00:33] everywhere I go. If you want to give it a try, download the SY app from the app
[01:00:36] a try, download the SY app from the app store now and scan the QR code on
[01:00:37] store now and scan the QR code on screen. And if you want 15% off your
[01:00:40] screen. And if you want 15% off your first purchase, use my code D
[01:00:44] first purchase, use my code D when you get to check out. That's D O A
[01:00:47] when you get to check out. That's D O A for 15% off. Keep that to yourself. My
[01:00:51] for 15% off. Keep that to yourself. My my car that drives itself that is AI
[01:00:54] my car that drives itself that is AI technology.
[01:00:54] technology. >> Yes.
[01:00:55] >> Yes. >> I sat here with Dra from Uber and he was
[01:00:57] >> I sat here with Dra from Uber and he was saying that I think in a couple of years
[01:00:59] saying that I think in a couple of years time
[01:01:01] time we won't need drivers um for Uber
[01:01:04] we won't need drivers um for Uber because the cars will drive themselves
[01:01:06] because the cars will drive themselves like they'll be fully autonomous.
[01:01:08] like they'll be fully autonomous. >> And I think if I'm not mistaken
[01:01:11] >> And I think if I'm not mistaken driving is one of the biggest
[01:01:12] driving is one of the biggest professions on planet earth. So when you
[01:01:15] professions on planet earth. So when you hear people when you hear these CEOs
[01:01:16] hear people when you hear these CEOs saying that there will be job disruption
[01:01:19] saying that there will be job disruption >> you say that they are not telling the
[01:01:22] >> you say that they are not telling the truth.
[01:01:22] truth. >> Yes. Or they're guessing in a way that's
[01:01:24] >> Yes. Or they're guessing in a way that's very good for them. Think about it from
[01:01:26] very good for them. Think about it from perspective of Microsoft Sachin Nadella.
[01:01:29] perspective of Microsoft Sachin Nadella. He's not going to be like yeah we don't
[01:01:30] He's not going to be like yeah we don't know if this is going to work mate. Of
[01:01:32] know if this is going to work mate. Of course he's going to talk his book and
[01:01:33] course he's going to talk his book and he's going to say yeah this is going to
[01:01:35] he's going to say yeah this is going to replace all workers. It's going to be
[01:01:36] replace all workers. It's going to be amazing. He it's going to be so
[01:01:38] amazing. He it's going to be so powerful. And then he'll change his tune
[01:01:40] powerful. And then he'll change his tune and say actually it's not going to
[01:01:41] and say actually it's not going to replace workers. that make him more
[01:01:42] replace workers. that make him more powerful because the things aren't
[01:01:44] powerful because the things aren't catching up. Dor from Uber for example,
[01:01:46] catching up. Dor from Uber for example, of course he's going to say if this
[01:01:48] of course he's going to say if this happens then that would be good for Uber
[01:01:50] happens then that would be good for Uber because Uber would just become an
[01:01:52] because Uber would just become an autonomous taxi service. There's a
[01:01:54] autonomous taxi service. There's a reason that Whimo's taken I I find Whimo
[01:01:56] reason that Whimo's taken I I find Whimo fascinating. I think that it's
[01:01:57] fascinating. I think that it's really cool. I think there are
[01:01:58] really cool. I think there are socioeconomic problems that will come
[01:01:59] socioeconomic problems that will come from it. I think there are actual real
[01:02:01] from it. I think there are actual real problems that will emerge and also
[01:02:03] problems that will emerge and also >> what kind of problems?
[01:02:04] >> what kind of problems? >> Well, I mean socioeconomically there are
[01:02:06] >> Well, I mean socioeconomically there are like you said one of the largest
[01:02:07] like you said one of the largest employment centers in the world. I mean
[01:02:09] employment centers in the world. I mean just the economics of cabs will fall
[01:02:11] just the economics of cabs will fall apart but again we are nowhere nowhere
[01:02:12] apart but again we are nowhere nowhere nowhere near that. We're not even close.
[01:02:14] nowhere near that. We're not even close. Whimo has had to do the smallest
[01:02:17] Whimo has had to do the smallest rollouts and the most control things
[01:02:18] rollouts and the most control things because the problem with pretty much
[01:02:20] because the problem with pretty much every AI system but especially driving
[01:02:22] every AI system but especially driving is not the getting 95% of the way. It's
[01:02:25] is not the getting 95% of the way. It's those edge cases. It's raining which is
[01:02:27] those edge cases. It's raining which is a big problem for them in San Francisco.
[01:02:29] a big problem for them in San Francisco. It's a kid runs across the road but
[01:02:31] It's a kid runs across the road but they're wearing a high viz thing. Does
[01:02:33] they're wearing a high viz thing. Does it even notice it's a child? Again, this
[01:02:35] it even notice it's a child? Again, this is a really interesting but very very
[01:02:36] is a really interesting but very very applicable example of uh the right
[01:02:38] applicable example of uh the right comparison to be made shouldn't be
[01:02:40] comparison to be made shouldn't be autonomous vehicles versus perfection.
[01:02:43] autonomous vehicles versus perfection. It should be autonomous vehicles versus
[01:02:44] It should be autonomous vehicles versus human drivers. I mean, I don't know if I
[01:02:47] human drivers. I mean, I don't know if I agree because a human driver might make
[01:02:48] agree because a human driver might make mistakes, sure, but again, not an expert
[01:02:51] mistakes, sure, but again, not an expert in autonomous cars. Just want to be
[01:02:53] in autonomous cars. Just want to be clear. But if we're pushing autonomous
[01:02:55] clear. But if we're pushing autonomous cars out there willy-nilly and we're not
[01:02:57] cars out there willy-nilly and we're not doing so in extremely controlled
[01:02:59] doing so in extremely controlled environments, those edge cases will
[01:03:01] environments, those edge cases will multiply and be dangerous. Yeah, they
[01:03:03] multiply and be dangerous. Yeah, they might be better at human drivers in some
[01:03:04] might be better at human drivers in some ways, but they might also I was in Vegas
[01:03:06] ways, but they might also I was in Vegas the other day and I was in a hotel and I
[01:03:08] the other day and I was in a hotel and I watched a bunch of Zuk's cars just get
[01:03:10] watched a bunch of Zuk's cars just get stuck.
[01:03:11] stuck. >> They're autonomous cars.
[01:03:12] >> They're autonomous cars. >> Yeah, they these weird boxy things. They
[01:03:14] >> Yeah, they these weird boxy things. They just blocked the exit. They just all
[01:03:16] just blocked the exit. They just all kind of lined up and just fell asleep. I
[01:03:19] kind of lined up and just fell asleep. I saw the same thing actually happen
[01:03:20] saw the same thing actually happen outside of a hotel when I got out of a
[01:03:21] outside of a hotel when I got out of a Whimo in San Francisco. Just stopped at
[01:03:23] Whimo in San Francisco. Just stopped at the and then a bunch of cars and another
[01:03:25] the and then a bunch of cars and another Whimo got stuck behind it. And these are
[01:03:27] Whimo got stuck behind it. And these are kind of
[01:03:27] kind of >> I've seen some human bad drivers as
[01:03:29] >> I've seen some human bad drivers as well. I I agree, but it's just we have
[01:03:31] well. I I agree, but it's just we have control over deploying these bad or good
[01:03:35] control over deploying these bad or good drivers. We have an ability to roll them
[01:03:37] drivers. We have an ability to roll them out slowly, which is exactly what we
[01:03:39] out slowly, which is exactly what we should do. I'm not saying autonomous
[01:03:41] should do. I'm not saying autonomous cars are bad. I'm saying we need to be
[01:03:43] cars are bad. I'm saying we need to be so so so careful and treat them as
[01:03:46] so so so careful and treat them as guilty and pro till proven innocent
[01:03:48] guilty and pro till proven innocent because we can prove and also they have
[01:03:51] because we can prove and also they have people overlooking them. They actually
[01:03:52] people overlooking them. They actually have people monitoring the roots. It is
[01:03:54] have people monitoring the roots. It is something they cannot rush out and it
[01:03:56] something they cannot rush out and it doesn't seem like they're rushing it,
[01:03:57] doesn't seem like they're rushing it, which is good. and they're not promising
[01:03:58] which is good. and they're not promising the world.
[01:03:59] the world. >> I do agree. Listen, I I'm a big fan of a
[01:04:01] >> I do agree. Listen, I I'm a big fan of a big fan of taxi drivers generally in
[01:04:03] big fan of taxi drivers generally in part because I spend a lot of time in
[01:04:04] part because I spend a lot of time in taxis and I think I'm not just getting
[01:04:05] taxis and I think I'm not just getting in there because I want to get to from A
[01:04:07] in there because I want to get to from A to B. I'm getting in there for lots of
[01:04:08] to B. I'm getting in there for lots of other reasons.
[01:04:09] other reasons. >> Yeah.
[01:04:09] >> Yeah. >> However, when I look at the stats
[01:04:12] >> However, when I look at the stats >> around what is more dangerous
[01:04:14] >> around what is more dangerous >> driving myself or having an autonomous
[01:04:16] >> driving myself or having an autonomous vehicle drive me, there's an 68% lower
[01:04:19] vehicle drive me, there's an 68% lower overall crash involvement rate when
[01:04:21] overall crash involvement rate when you're in an an autonomous vehicle. Mhm.
[01:04:24] you're in an an autonomous vehicle. Mhm. >> Autonomous vehicles experience roughly
[01:04:25] >> Autonomous vehicles experience roughly 2.1 police reported crashes per million
[01:04:28] 2.1 police reported crashes per million miles compared to humans that are at
[01:04:29] miles compared to humans that are at roughly 4.68 per million miles. So, a
[01:04:32] roughly 4.68 per million miles. So, a 55% reduction when you get in an
[01:04:34] 55% reduction when you get in an autonomous vehicle. And autonomous
[01:04:36] autonomous vehicle. And autonomous vehicles show an 80 to 81% reduction in
[01:04:39] vehicles show an 80 to 81% reduction in crashes resulting in injuries versus
[01:04:41] crashes resulting in injuries versus human drivers.
[01:04:42] human drivers. >> Uhhuh.
[01:04:43] >> Uhhuh. >> So, you're 85% less likely to be
[01:04:45] >> So, you're 85% less likely to be involved in a single vehicle crash like
[01:04:48] involved in a single vehicle crash like hitting a wall or a tree if you're an
[01:04:50] hitting a wall or a tree if you're an autonomous vehicle
[01:04:51] autonomous vehicle >> versus being driven by I agree. But
[01:04:54] >> versus being driven by I agree. But >> so it's safer
[01:04:55] >> so it's safer >> in also that data is what's the sample
[01:04:59] >> in also that data is what's the sample size of human drivers? I mean we've got
[01:05:01] size of human drivers? I mean we've got many many many many many many more years
[01:05:03] many many many many many many more years of drivers and many many many more years
[01:05:04] of drivers and many many many more years of accidents and also man does that not
[01:05:07] of accidents and also man does that not have anything to do with generative AI.
[01:05:09] have anything to do with generative AI. If we were just talking about that be
[01:05:11] If we were just talking about that be having a different conversation.
[01:05:12] having a different conversation. >> I guess the question here was really
[01:05:13] >> I guess the question here was really around job disruption. Like you know we
[01:05:15] around job disruption. Like you know we we look across industries and we go
[01:05:17] we look across industries and we go driving is a massive profession. Is
[01:05:18] driving is a massive profession. Is there going to be job disruption because
[01:05:20] there going to be job disruption because cars can now drive themselves? If we
[01:05:22] cars can now drive themselves? If we think about white collar, you know,
[01:05:23] think about white collar, you know, jobs, you know, lawyers and accountants,
[01:05:26] jobs, you know, lawyers and accountants, people sit here and they tell me that
[01:05:27] people sit here and they tell me that lawyers and accountants would the
[01:05:30] lawyers and accountants would the profession, right? I should say some of
[01:05:31] profession, right? I should say some of the skills within the profession will be
[01:05:33] the skills within the profession will be relegated to AIS to do.
[01:05:36] relegated to AIS to do. >> Here's the thing. Lawyers, for example,
[01:05:38] >> Here's the thing. Lawyers, for example, great example. Always hearing
[01:05:40] great example. Always hearing legal partners talking about AI. Never
[01:05:43] legal partners talking about AI. Never the associates. The associates are the
[01:05:45] the associates. The associates are the ones that go out and find the president.
[01:05:46] ones that go out and find the president. They're the ones that go and do the
[01:05:48] They're the ones that go and do the grunt work. They're the ones who are
[01:05:49] grunt work. They're the ones who are pulling motions half the time. The
[01:05:50] pulling motions half the time. The partner is the one that might be the
[01:05:52] partner is the one that might be the litigant. It may be the client facing,
[01:05:53] litigant. It may be the client facing, but the ones that are actually doing the
[01:05:55] but the ones that are actually doing the day-to-day work. I'm not hearing from
[01:05:56] day-to-day work. I'm not hearing from them. I'm not hearing associates being
[01:05:58] them. I'm not hearing associates being like, "This is awesome." I'm
[01:06:00] like, "This is awesome." I'm hearing a bunch of well- paid people
[01:06:02] hearing a bunch of well- paid people that have sat on Chat GPT and gone,
[01:06:04] that have sat on Chat GPT and gone, "Yeah, yeah, I'm the greatest lawyer
[01:06:06] "Yeah, yeah, I'm the greatest lawyer ever." They're not the ones that I want
[01:06:08] ever." They're not the ones that I want to hear from the actual workers. White
[01:06:10] to hear from the actual workers. White collar labor disruption is not
[01:06:11] collar labor disruption is not happening. Open AAI had a study that
[01:06:13] happening. Open AAI had a study that came out I think like a week ago that
[01:06:15] came out I think like a week ago that said there was no corre connection
[01:06:17] said there was no corre connection between spending on AI tokens and
[01:06:19] between spending on AI tokens and revenue per employee. Like this is open
[01:06:21] revenue per employee. Like this is open and that's
[01:06:22] and that's >> what does that mean? Could you explain
[01:06:23] >> what does that mean? Could you explain that to me?
[01:06:23] that to me? >> As in the more tokens you spend has no
[01:06:26] >> As in the more tokens you spend has no no correlation at all with the amount of
[01:06:29] no correlation at all with the amount of money you make. It's the second report
[01:06:30] money you make. It's the second report they've put out. The other one was like
[01:06:32] they've put out. The other one was like hallucinations are mathematically
[01:06:33] hallucinations are mathematically guaranteed kind of almost the one thing
[01:06:36] guaranteed kind of almost the one thing I respect about that company that
[01:06:37] I respect about that company that occasion they just put out a study. It's
[01:06:38] occasion they just put out a study. It's like, yeah, kind of sucks.
[01:06:41] like, yeah, kind of sucks. [clears throat] But the people that are
[01:06:42] [clears throat] But the people that are having their lives disrupted work-wise
[01:06:45] having their lives disrupted work-wise are art directors. It's people, art
[01:06:47] are art directors. It's people, art directors, transcribers, translators,
[01:06:50] directors, transcribers, translators, who have bosses that don't care about
[01:06:52] who have bosses that don't care about the output. It's what they consider
[01:06:54] the output. It's what they consider cheap work. And the problem is is those
[01:06:56] cheap work. And the problem is is those people would have automated your work
[01:06:57] people would have automated your work away anyway. They would have sold it.
[01:06:59] away anyway. They would have sold it. They would have taken the cheapest for
[01:07:00] They would have taken the cheapest for they would have sold it to the global
[01:07:02] they would have sold it to the global self. They would have taken the
[01:07:03] self. They would have taken the shittiest option they could. That is
[01:07:04] shittiest option they could. That is something that AI is doing. And again,
[01:07:06] something that AI is doing. And again, those people are not paying the actual
[01:07:07] those people are not paying the actual cost of AI. They're using a
[01:07:09] cost of AI. They're using a subscription. The actual white collar
[01:07:11] subscription. The actual white collar labor force might have some things that
[01:07:15] labor force might have some things that are slightly changing, but there is no
[01:07:17] are slightly changing, but there is no evidence of like productivity gains. In
[01:07:20] evidence of like productivity gains. In fact, if there were, they would be
[01:07:22] fact, if there were, they would be screaming it from the rooftops. There
[01:07:23] screaming it from the rooftops. There was an Oxford economics study last year
[01:07:25] was an Oxford economics study last year where it's like, oh, young people are
[01:07:27] where it's like, oh, young people are finding less jobs because of AI. We
[01:07:29] finding less jobs because of AI. We actually read the study, which multiple
[01:07:31] actually read the study, which multiple journalists did not. It was a single
[01:07:33] journalists did not. It was a single line that said, "Yeah, we saw some
[01:07:34] line that said, "Yeah, we saw some correlation." Didn't give a number.
[01:07:37] correlation." Didn't give a number. Didn't actually say what the correlation
[01:07:39] Didn't actually say what the correlation was. We are so conditioned to believe
[01:07:41] was. We are so conditioned to believe that the rich and powerful know what
[01:07:43] that the rich and powerful know what they're doing that we internalize these
[01:07:46] they're doing that we internalize these narratives about like, well, previous
[01:07:48] narratives about like, well, previous booms lost a lot of money. Well,
[01:07:50] booms lost a lot of money. Well, technology takes time to do stuff. And
[01:07:52] technology takes time to do stuff. And they are intentionally playing on those
[01:07:55] they are intentionally playing on those mythologies. They are playing on these
[01:07:57] mythologies. They are playing on these knowing that journalists, analysts,
[01:07:59] knowing that journalists, analysts, investors will believe them. And this is
[01:08:02] investors will believe them. And this is partly because our our realities are
[01:08:04] partly because our our realities are defined by stock prices. Because the
[01:08:06] defined by stock prices. Because the stock prices of these companies went up,
[01:08:07] stock prices of these companies went up, we're like, "Oh, look, it must be
[01:08:09] we're like, "Oh, look, it must be working, right?"
[01:08:10] working, right?" >> Both of those things you said were true,
[01:08:12] >> Both of those things you said were true, though, right? Like that previous
[01:08:13] though, right? Like that previous technologies didn't make money at the
[01:08:14] technologies didn't make money at the start and you The other one you said was
[01:08:16] start and you The other one you said was um they'll get better.
[01:08:18] um they'll get better. >> But that's the thing. Okay. Because
[01:08:20] >> But that's the thing. Okay. Because another thing got better, this will get
[01:08:21] another thing got better, this will get better.
[01:08:22] better. >> No, but there's there's got to be
[01:08:23] >> No, but there's there's got to be something that they're saying that is
[01:08:24] something that they're saying that is fundamentally not true because those are
[01:08:26] fundamentally not true because those are two true statements that okay,
[01:08:27] two true statements that okay, technology often starts
[01:08:29] technology often starts >> I know. I get what you mean. What they
[01:08:30] >> I know. I get what you mean. What they are fundamentally misleading people
[01:08:32] are fundamentally misleading people about is how possible it is. How many
[01:08:34] about is how possible it is. How many actual signs they have because they
[01:08:36] actual signs they have because they don't have the signs. If they had the
[01:08:37] don't have the signs. If they had the signs as in the signs of this getting
[01:08:39] signs as in the signs of this getting cheaper as in the signs of this being
[01:08:41] cheaper as in the signs of this being able to autonomously do work without the
[01:08:43] able to autonomously do work without the Rub Goldberg machine and even then in a
[01:08:45] Rub Goldberg machine and even then in a reliable way that was making the
[01:08:47] reliable way that was making the customer more money being productive in
[01:08:50] customer more money being productive in a way you can say with your whole chest
[01:08:51] a way you can say with your whole chest without a series of asterisks and that's
[01:08:54] without a series of asterisks and that's how it is across the board. The people
[01:08:57] how it is across the board. The people that are most excited about this,
[01:08:59] that are most excited about this, psychopaths on Twitter in many cases are
[01:09:01] psychopaths on Twitter in many cases are people that I believe there really are
[01:09:04] people that I believe there really are some I'm sorry, there are some people on
[01:09:06] some I'm sorry, there are some people on Twitter because the other thing about
[01:09:07] Twitter because the other thing about this is this is really unique to the AI
[01:09:10] this is this is really unique to the AI industry. I've never seen it any other
[01:09:11] industry. I've never seen it any other industry outside of maybe like sports
[01:09:13] industry outside of maybe like sports teams. The attachment that some people
[01:09:15] teams. The attachment that some people online have to these companies. If you
[01:09:18] online have to these companies. If you dare dare to criticize anthropic, it's
[01:09:21] dare dare to criticize anthropic, it's almost this religious attachment. Good
[01:09:23] almost this religious attachment. Good example was this week Bloomberg reported
[01:09:25] example was this week Bloomberg reported that OpenAI was on track to hit $40
[01:09:28] that OpenAI was on track to hit $40 billion in annualized revenue. Month
[01:09:30] billion in annualized revenue. Month times 12, four weeks times 13, we don't
[01:09:31] times 12, four weeks times 13, we don't know. They don't define it. I saw
[01:09:34] know. They don't define it. I saw multiple people and I going actually
[01:09:35] multiple people and I going actually it's 60 billion. It's actually 60
[01:09:37] it's 60 billion. It's actually 60 billion. I heard from someone it is like
[01:09:40] billion. I heard from someone it is like a cult and it's a cult of software
[01:09:43] a cult and it's a cult of software driven around growth and this idea that
[01:09:46] driven around growth and this idea that by backing the right horse you will have
[01:09:49] by backing the right horse you will have some grand thing and open AI in
[01:09:52] some grand thing and open AI in particular in particular Mr. Baltman
[01:09:55] particular in particular Mr. Baltman they have been fermenting this that Tibo
[01:09:57] they have been fermenting this that Tibo as well the Tibbo the one of the guys at
[01:10:00] as well the Tibbo the one of the guys at uh OpenAI they ferment this thing online
[01:10:02] uh OpenAI they ferment this thing online they build this kind of parasocial
[01:10:04] they build this kind of parasocial relationship with both the large
[01:10:06] relationship with both the large language model themselves and the
[01:10:07] language model themselves and the companies and one's allegiance to the
[01:10:10] companies and one's allegiance to the companies is so important it's truly
[01:10:12] companies is so important it's truly vile if only these people gave a
[01:10:15] vile if only these people gave a about I don't know Medicare for all or
[01:10:17] about I don't know Medicare for all or poverty or thing like actual problems in
[01:10:20] poverty or thing like actual problems in the world versus are we buying enough
[01:10:22] the world versus are we buying enough GPUs Do you know what's interesting is
[01:10:24] GPUs Do you know what's interesting is some of what your narrative
[01:10:27] some of what your narrative one would argue actually helps them.
[01:10:29] one would argue actually helps them. How? Because you know the AI doomers
[01:10:32] How? Because you know the AI doomers that have come here and told you know
[01:10:33] that have come here and told you know some of the original founding fathers of
[01:10:34] some of the original founding fathers of AI like Jeffrey Hinton have told me that
[01:10:38] AI like Jeffrey Hinton have told me that what they're building is highly highly
[01:10:39] what they're building is highly highly dangerous and that it will be
[01:10:40] dangerous and that it will be fundamentally disruptive to society. And
[01:10:44] fundamentally disruptive to society. And it's interesting because some of the
[01:10:45] it's interesting because some of the CEOs who you've mentioned, their
[01:10:47] CEOs who you've mentioned, their historical narrative was also, by the
[01:10:49] historical narrative was also, by the way, this is really dangerous
[01:10:50] way, this is really dangerous and there is a significant chance it
[01:10:51] and there is a significant chance it could f we could up the planet.
[01:10:53] could f we could up the planet. >> And what we've seen is this slow pivot
[01:10:56] >> And what we've seen is this slow pivot away from it because now they're getting
[01:10:58] away from it because now they're getting booed and they're being attacked.
[01:10:59] booed and they're being attacked. There've been this slow pivot away from
[01:11:01] There've been this slow pivot away from it. And the pivot almost sounds a little
[01:11:04] it. And the pivot almost sounds a little bit like your narrative.
[01:11:06] bit like your narrative. >> It now sounds like actually no, it's not
[01:11:08] >> It now sounds like actually no, it's not going to change anything and you're all
[01:11:09] going to change anything and you're all going to be fine. And it's now there's
[01:11:10] going to be fine. And it's now there's just not it's nah it's not dangerous at
[01:11:12] just not it's nah it's not dangerous at all.
[01:11:12] all. >> But that's the funny thing
[01:11:13] >> But that's the funny thing >> and that's why I'm saying like you're
[01:11:15] >> and that's why I'm saying like you're you're not they I actually think there
[01:11:17] you're not they I actually think there might be a couple PR people at these big
[01:11:18] might be a couple PR people at these big AI companies thinking thank god for Ed
[01:11:22] AI companies thinking thank god for Ed some [laughter] of it because you're
[01:11:24] some [laughter] of it because you're like you're saying actually don't worry
[01:11:25] like you're saying actually don't worry everything's going to be fine. It's not
[01:11:27] everything's going to be fine. It's not going to take your job. It's not going
[01:11:28] going to take your job. It's not going to disrupt the economy. It's just a fad.
[01:11:29] to disrupt the economy. It's just a fad. There's no technology. And I think they
[01:11:31] There's no technology. And I think they don't think that.
[01:11:32] don't think that. >> Here's the thing. I think Alman and
[01:11:33] >> Here's the thing. I think Alman and Amday are some of the most deeply
[01:11:35] Amday are some of the most deeply corrupt and cynical people in the world.
[01:11:36] corrupt and cynical people in the world. I don't think of course they were going
[01:11:37] I don't think of course they were going to say from the it was early 2023 or man
[01:11:41] to say from the it was early 2023 or man said we're a little bit scared about
[01:11:42] said we're a little bit scared about what we're creating. Oh, shut up. I'm
[01:11:45] what we're creating. Oh, shut up. I'm just I hear that and I feel so
[01:11:46] just I hear that and I feel so frustrated because I've met so many of
[01:11:48] frustrated because I've met so many of these rich liars, these people.
[01:11:51] these rich liars, these people. And you know why he wants to say that?
[01:11:52] And you know why he wants to say that? So you'll invest in his company and buy
[01:11:54] So you'll invest in his company and buy the software. So you'll be scared that
[01:11:56] the software. So you'll be scared that if you don't use AI today, you'll be
[01:11:58] if you don't use AI today, you'll be left behind in the future, which is
[01:11:59] left behind in the future, which is their continual narrative that if you
[01:12:01] their continual narrative that if you don't get on the train today,
[01:12:04] don't get on the train today, then you'll be left behind. By the way,
[01:12:05] then you'll be left behind. By the way, every single scam and con starts with
[01:12:08] every single scam and con starts with rushing you. Every single trick in
[01:12:10] rushing you. Every single trick in history begins with saying you must do
[01:12:12] history begins with saying you must do this now. And best piece of advice I
[01:12:14] this now. And best piece of advice I ever got was if anyone tries to rush you
[01:12:17] ever got was if anyone tries to rush you and it's not literally a mortal thing
[01:12:18] and it's not literally a mortal thing like you are bleeding or on fire or the
[01:12:20] like you are bleeding or on fire or the house is on fire, slow down. And yet all
[01:12:23] house is on fire, slow down. And yet all of these companies saying it's so scary.
[01:12:25] of these companies saying it's so scary. And now they're talking about slowdowns.
[01:12:27] And now they're talking about slowdowns. But you ever noticed that Amade and
[01:12:28] But you ever noticed that Amade and Ortman, they say, "Oh, maybe we should
[01:12:30] Ortman, they say, "Oh, maybe we should slow down progress." And then they
[01:12:31] slow down progress." And then they don't. Right now, Orman's saying, "Oh,
[01:12:33] don't. Right now, Orman's saying, "Oh, we slow down progress because we're so
[01:12:35] we slow down progress because we're so delayed." No, they're out of compute.
[01:12:36] delayed." No, they're out of compute. Now, they're doing it. I can guarantee
[01:12:38] Now, they're doing it. I can guarantee you, by the way, their PR people do not
[01:12:39] you, by the way, their PR people do not like me. I know for I know I don't think
[01:12:41] like me. I know for I know I don't think OpenAI's PR people are super fond of me.
[01:12:43] OpenAI's PR people are super fond of me. >> But I bet there's elements of what
[01:12:45] >> But I bet there's elements of what you're saying because you're calming
[01:12:46] you're saying because you're calming people. You You are theoretically
[01:12:48] people. You You are theoretically calming down the general public.
[01:12:49] calming down the general public. >> And you know what? I hope I am because
[01:12:52] >> And you know what? I hope I am because >> the fear based tactics is horrible.
[01:12:53] >> the fear based tactics is horrible. These companies don't want that. These
[01:12:55] These companies don't want that. These companies want people scared. I'm 100%
[01:12:57] companies want people scared. I'm 100% sure.
[01:12:57] sure. >> Uh I don't I just fundamentally
[01:12:59] >> Uh I don't I just fundamentally disagree. I think it
[01:13:00] disagree. I think it >> can I so the timelines there and I sit
[01:13:02] >> can I so the timelines there and I sit here and what I do is I log their quotes
[01:13:05] here and what I do is I log their quotes over time
[01:13:06] over time >> and I read them out from 2015
[01:13:08] >> and I read them out from 2015 >> to 2026 and the change you see is them
[01:13:12] >> to 2026 and the change you see is them going from there could be extinction
[01:13:15] going from there could be extinction that's the narrative the early narrative
[01:13:16] that's the narrative the early narrative Elon said it himself he says it's the
[01:13:17] Elon said it himself he says it's the single most dangerous thing in
[01:13:19] single most dangerous thing in >> Elon and then you track it over time and
[01:13:21] >> Elon and then you track it over time and it evolves to this age of abundance
[01:13:24] it evolves to this age of abundance we're all going to have unlimited stuff
[01:13:26] we're all going to have unlimited stuff and then um the the new slogan at
[01:13:28] and then um the the new slogan at trackbt is intelligence for everyone.
[01:13:31] trackbt is intelligence for everyone. It's suddenly and all the and and
[01:13:32] It's suddenly and all the and and whenever Daario comes out and says, "By
[01:13:34] whenever Daario comes out and says, "By the way, it's really dangerous."
[01:13:36] the way, it's really dangerous." They attack Daario. Yeah. They hate him.
[01:13:39] They attack Daario. Yeah. They hate him. >> That man [laughter] Daario is
[01:13:40] >> That man [laughter] Daario is >> They're like, "Dario, shut the up."
[01:13:42] >> They're like, "Dario, shut the up." >> Honestly, I I've been saying Dario, shut
[01:13:44] >> Honestly, I I've been saying Dario, shut the up for years. But it's But the
[01:13:46] the up for years. But it's But the thing is, I get your point where it's
[01:13:47] thing is, I get your point where it's like I don't think they've changed to
[01:13:50] like I don't think they've changed to calm the public down so much as they're
[01:13:51] calm the public down so much as they're desperate to not get regulated, which is
[01:13:53] desperate to not get regulated, which is laughable. We don't regulate tech. We
[01:13:56] laughable. We don't regulate tech. We don't regulate America doesn't
[01:13:58] don't regulate America doesn't regulate We are in the We are
[01:14:00] regulate We are in the We are still trapped in the hands of Milton
[01:14:03] still trapped in the hands of Milton Freriedman, Margaret Thatcher, and
[01:14:04] Freriedman, Margaret Thatcher, and Ronald Reagan. We're still stuck
[01:14:06] Ronald Reagan. We're still stuck in the neoliberalistic hellscape, which
[01:14:09] in the neoliberalistic hellscape, which is growth at all cost, free market
[01:14:11] is growth at all cost, free market capitalism. So, no, no one's regulating
[01:14:14] capitalism. So, no, no one's regulating the regulation of these companies should
[01:14:15] the regulation of these companies should have been, I don't know, breaking up.
[01:14:18] have been, I don't know, breaking up. Put these bastards to the side. Break up
[01:14:20] Put these bastards to the side. Break up these for sure. We shouldn't
[01:14:21] these for sure. We shouldn't have companies this big. It makes things
[01:14:22] have companies this big. It makes things worse.
[01:14:23] worse. >> But these technologies are dangerous.
[01:14:25] >> But these technologies are dangerous. >> I mean, they're dangerous, but not in
[01:14:26] >> I mean, they're dangerous, but not in the ways they've been warning about.
[01:14:28] the ways they've been warning about. Let's if we think about cyber hacking,
[01:14:30] Let's if we think about cyber hacking, >> right? And just to be clear, those cyber
[01:14:32] >> right? And just to be clear, those cyber hacking things that happened were not a
[01:14:34] hacking things that happened were not a result of they were like break out of
[01:14:36] result of they were like break out of the sandbox and then they set the
[01:14:37] the sandbox and then they set the sandbox up wrong. They set up the server
[01:14:40] sandbox up wrong. They set up the server they were on wrong. But I mean, you
[01:14:41] they were on wrong. But I mean, you know, advanced AI models could very
[01:14:43] know, advanced AI models could very easily cuz they can go out onto the open
[01:14:45] easily cuz they can go out onto the open internet as agents. They could very
[01:14:47] internet as agents. They could very easily go and look at code bases of
[01:14:49] easily go and look at code bases of different websites, find vulnerabilities
[01:14:51] different websites, find vulnerabilities and exploit those vulnerabilities.
[01:14:52] and exploit those vulnerabilities. >> Yeah. in at scale and arguably um at a
[01:14:56] >> Yeah. in at scale and arguably um at a higher intelligence and faster and wider
[01:14:59] higher intelligence and faster and wider than humans a human hacker could
[01:15:01] than humans a human hacker could theoretically. So that's dangerous.
[01:15:03] theoretically. So that's dangerous. >> Well, here's the funny thing. We don't
[01:15:05] >> Well, here's the funny thing. We don't know how much compute was spent to do
[01:15:07] know how much compute was spent to do the hugging face attack, the open AI
[01:15:09] the hugging face attack, the open AI one. We also do know that they
[01:15:10] one. We also do know that they improperly set up the server to keep it
[01:15:12] improperly set up the server to keep it in. They thought they'd turn the
[01:15:14] in. They thought they'd turn the internet off and they didn't. That's
[01:15:15] internet off and they didn't. That's human error. And that's human error in a
[01:15:17] human error. And that's human error in a sense that yeah, they threw about an
[01:15:19] sense that yeah, they threw about an indeterminately large amount of compute.
[01:15:21] indeterminately large amount of compute. This is dangerous, but people keep
[01:15:23] This is dangerous, but people keep saying we can't let the the Chinese get
[01:15:26] saying we can't let the the Chinese get a hold of these models. We couldn't
[01:15:27] a hold of these models. We couldn't possibly because what if these models
[01:15:29] possibly because what if these models fall into the wrong hands? They're
[01:15:30] fall into the wrong hands? They're already in the wrong hands. Mark
[01:15:32] already in the wrong hands. Mark Zuckerberg, Sam Olman, Dario Amade. The
[01:15:35] Zuckerberg, Sam Olman, Dario Amade. The wrong hands are the hands of those who
[01:15:37] wrong hands are the hands of those who are running these companies. We should
[01:15:39] are running these companies. We should not be training these models to do these
[01:15:42] not be training these models to do these things. I don't know why the we're
[01:15:43] things. I don't know why the we're doing it other than they've run out of
[01:15:45] doing it other than they've run out of other things they can train on. There's
[01:15:47] other things they can train on. There's a ton. And the fact that they can do it,
[01:15:49] a ton. And the fact that they can do it, it's kind of interesting. But you do
[01:15:50] it's kind of interesting. But you do would you agree that it's an
[01:15:53] would you agree that it's an intelligence and I'll call it that you
[01:15:55] intelligence and I'll call it that you know you might disagree with that
[01:15:56] know you might disagree with that terminology but an intelligence that can
[01:15:57] terminology but an intelligence that can go out onto the internet and click
[01:16:00] go out onto the internet and click around and take actions is inherently
[01:16:04] around and take actions is inherently there's risks associated with that. Well
[01:16:06] there's risks associated with that. Well the second part I agree with the risks
[01:16:08] the second part I agree with the risks we've had people running automated
[01:16:10] we've had people running automated scripts hacking scripts for a while
[01:16:12] scripts hacking scripts for a while we've had hackers doing that for years
[01:16:13] we've had hackers doing that for years and years and years. This is brute
[01:16:15] and years and years. This is brute forcing it with a bunch of compute and
[01:16:16] forcing it with a bunch of compute and yet it is dangerous. These companies are
[01:16:19] yet it is dangerous. These companies are doing something dangerous. That is not
[01:16:21] doing something dangerous. That is not what Jeffrey Hinton at have been warning
[01:16:24] what Jeffrey Hinton at have been warning about. They've been saying, "Oh, these
[01:16:25] about. They've been saying, "Oh, these things could destroy society. They could
[01:16:27] things could destroy society. They could manipulate people." When you actually
[01:16:28] manipulate people." When you actually look at the underlying things, not so
[01:16:30] look at the underlying things, not so much. Jeffrey Hinton as well talking his
[01:16:32] much. Jeffrey Hinton as well talking his book still got his Google stock, I
[01:16:33] book still got his Google stock, I think. And weirdly enough, he left
[01:16:34] think. And weirdly enough, he left Google because he was worried about the
[01:16:35] Google because he was worried about the AI there, but then immediately made a
[01:16:38] AI there, but then immediately made a comment being like, "Yeah, actually
[01:16:39] comment being like, "Yeah, actually though, Google's very responsible."
[01:16:41] though, Google's very responsible." Strange thing. But let's get back to the
[01:16:42] Strange thing. But let's get back to the the cyber security side. I agree this is
[01:16:45] the cyber security side. I agree this is dangerous. These people should not have
[01:16:46] dangerous. These people should not have access to so much comput. They clearly
[01:16:48] access to so much comput. They clearly don't know what to do with it. There's a
[01:16:50] don't know what to do with it. There's a really easy way of dealing with this.
[01:16:51] really easy way of dealing with this. It's not letting them use so much
[01:16:53] It's not letting them use so much compute. It's regulating that part out
[01:16:55] compute. It's regulating that part out of existence. What if the Chinese do it?
[01:16:57] of existence. What if the Chinese do it? The Chinese were able to distill the
[01:16:59] The Chinese were able to distill the models. And also,
[01:17:02] models. And also, I don't know, regulate it and stop I I
[01:17:05] I don't know, regulate it and stop I I feel like with this particular thing as
[01:17:07] feel like with this particular thing as well, we got to this point and let the
[01:17:10] well, we got to this point and let the genie out of the bottle to use an
[01:17:11] genie out of the bottle to use an annoying Samman term. We let this happen
[01:17:14] annoying Samman term. We let this happen because we let these companies be
[01:17:16] because we let these companies be unregulated and use as much computers we
[01:17:17] unregulated and use as much computers we want. We had these enablers
[01:17:19] want. We had these enablers allowing them to burn as much computers
[01:17:21] allowing them to burn as much computers as they want. And also we for all of
[01:17:24] as they want. And also we for all of these dire warnings about AI dangers, no
[01:17:27] these dire warnings about AI dangers, no one seems to have done anything.
[01:17:28] one seems to have done anything. >> Okay, we're going to play a game, Ed.
[01:17:30] >> Okay, we're going to play a game, Ed. >> Let's play it.
[01:17:30] >> Let's play it. >> On these cards here,
[01:17:32] >> On these cards here, >> I have the things that you consider to
[01:17:34] >> I have the things that you consider to be myths about the AI industry.
[01:17:37] be myths about the AI industry. >> The challenge is I want you to give me
[01:17:40] >> The challenge is I want you to give me one sentence.
[01:17:41] one sentence. on each myth.
[01:17:43] on each myth. >> Oh, Christ.
[01:17:43] >> Oh, Christ. >> So, just your first reaction. You're
[01:17:44] >> So, just your first reaction. You're going to pick it up, you're going to
[01:17:45] going to pick it up, you're going to read it,
[01:17:46] read it, >> and then you're going to give me one
[01:17:47] >> and then you're going to give me one sentence on your opinion of that
[01:17:49] sentence on your opinion of that >> um belief.
[01:17:50] >> um belief. >> Okay, let's go.
[01:17:51] >> Okay, let's go. >> So, let's do this.
[01:18:00] >> What does it say in your says the the AI industry is creating enormous economic
[01:18:02] industry is creating enormous economic growth?
[01:18:02] growth? >> No, it's not. It's nowhere in the data.
[01:18:06] >> No, it's not. It's nowhere in the data. >> Okay. [laughter] Like, it's just May I
[01:18:07] >> Okay. [laughter] Like, it's just May I do a second sentence?
[01:18:08] do a second sentence? >> Go ahead. pretty much all of the
[01:18:10] >> Go ahead. pretty much all of the economics is either Nvidia feeding money
[01:18:12] economics is either Nvidia feeding money to it companies like Corewave or these
[01:18:15] to it companies like Corewave or these three companies feeding money to these
[01:18:16] three companies feeding money to these ones to spend it with the them.
[01:18:18] ones to spend it with the them. >> Okay. And what evidence do you have that
[01:18:21] >> Okay. And what evidence do you have that there's it's not causing economic
[01:18:23] there's it's not causing economic growth?
[01:18:23] growth? >> Just to be clear, other than the spend
[01:18:25] >> Just to be clear, other than the spend on semiconductors, so the speculative
[01:18:27] on semiconductors, so the speculative investment in GPUs and data center
[01:18:29] investment in GPUs and data center infrastructure that's happening, but as
[01:18:31] infrastructure that's happening, but as far as like spend on AI goes, barely
[01:18:34] far as like spend on AI goes, barely cracking hundred billion. And most of
[01:18:36] cracking hundred billion. And most of that is just these two running their
[01:18:38] that is just these two running their services and paying these three
[01:18:39] services and paying these three companies, Oracle, Core, and others.
[01:18:41] companies, Oracle, Core, and others. >> But a hundred billion is a lot of money
[01:18:44] >> But a hundred billion is a lot of money for a relatively new technology.
[01:18:46] for a relatively new technology. >> Not when you've spent $300 billion in
[01:18:48] >> Not when you've spent $300 billion in equity funding. And it if we're going
[01:18:50] equity funding. And it if we're going with just these three, I think $600
[01:18:52] with just these three, I think $600 billion in capital expenditures.
[01:18:54] billion in capital expenditures. >> Yeah, I get that. That means it's not
[01:18:55] >> Yeah, I get that. That means it's not profitable. But the hundred billion is
[01:18:57] profitable. But the hundred billion is an expression of consumer demand
[01:18:59] an expression of consumer demand >> when the compute is mostly driven by
[01:19:01] >> when the compute is mostly driven by subscriptions that subsidized. No, it's
[01:19:03] subscriptions that subsidized. No, it's not. When you're giving someone $20 or
[01:19:04] not. When you're giving someone $20 or $40 for a dollar, they're going to use
[01:19:06] $40 for a dollar, they're going to use it more. If this was all on a per
[01:19:08] it more. If this was all on a per million token basis, we'd be having a
[01:19:10] million token basis, we'd be having a different conversation.
[01:19:10] different conversation. >> Okay, fair. Fine. Cool. Next one.
[01:19:14] >> Okay, fair. Fine. Cool. Next one. >> The United States need to spend
[01:19:16] >> The United States need to spend trillions to beat China in the AI race.
[01:19:19] trillions to beat China in the AI race. Let's see.
[01:19:21] Let's see. What AI race?
[01:19:24] What AI race? That's actually That's actually my
[01:19:25] That's actually That's actually my point. It's what AI race is there. Is it
[01:19:28] point. It's what AI race is there. Is it to make big scary LLMs? They they did
[01:19:30] to make big scary LLMs? They they did that already without the Nvidia GPUs. By
[01:19:33] that already without the Nvidia GPUs. By the way, they've got Blackwell GPUs.
[01:19:34] the way, they've got Blackwell GPUs. Kakashi and Jastario, two amazing
[01:19:36] Kakashi and Jastario, two amazing analysts I love. They've been on this
[01:19:38] analysts I love. They've been on this for years. It's like China's already had
[01:19:40] for years. It's like China's already had Nvidia GPUs that they're not meant to
[01:19:42] Nvidia GPUs that they're not meant to have for years. But also to do what?
[01:19:44] have for years. But also to do what? They already got the LMS. What What's
[01:19:46] They already got the LMS. What What's the race to do? To make us spend more
[01:19:47] the race to do? To make us spend more money than them? For us to constantly
[01:19:49] money than them? For us to constantly piss our pants worrying about China?
[01:19:51] piss our pants worrying about China? Because u they won if that's the case.
[01:19:54] Because u they won if that's the case. Myth number three, AI will replace all
[01:19:56] Myth number three, AI will replace all human jobs.
[01:19:58] human jobs. that just isn't happening and there's no
[01:20:00] that just isn't happening and there's no economic data to support it.
[01:20:02] economic data to support it. >> Will it replace some jobs?
[01:20:05] >> Will it replace some jobs? >> I mean, it's replaced some contract
[01:20:07] >> I mean, it's replaced some contract labor that would otherwise be replaced
[01:20:08] labor that would otherwise be replaced with cheap labor out in the global
[01:20:09] with cheap labor out in the global south. It's a digital globalization in
[01:20:12] south. It's a digital globalization in that sense, but all jobs, most jobs, a
[01:20:15] that sense, but all jobs, most jobs, a lot of jobs. No.
[01:20:16] lot of jobs. No. >> What about robotics?
[01:20:18] >> What about robotics? >> Robotics is not what we're talking
[01:20:19] >> Robotics is not what we're talking about. Robotics is a very different
[01:20:21] about. Robotics is a very different thing. And even then,
[01:20:21] thing. And even then, >> robotics will be powered by AI.
[01:20:23] >> robotics will be powered by AI. >> I mean, yes, but there are tons of
[01:20:25] >> I mean, yes, but there are tons of different kinds of AI. We're talking
[01:20:26] different kinds of AI. We're talking explicitly about generative AI. And
[01:20:28] explicitly about generative AI. And that's what I this mythbusters piece
[01:20:30] that's what I this mythbusters piece that was definitely about generative AI.
[01:20:31] that was definitely about generative AI. >> Okay. But what about robotics? Like the
[01:20:33] >> Okay. But what about robotics? Like the thing is the Optimus robot that Elon's
[01:20:35] thing is the Optimus robot that Elon's working on at Tesla.
[01:20:37] working on at Tesla. >> The one where even in the demo of the
[01:20:39] >> The one where even in the demo of the hand he like they had to have a guy
[01:20:41] hand he like they had to have a guy controlling it. Wasn't doing it
[01:20:42] controlling it. Wasn't doing it autonomously. Here's the thing. If they
[01:20:45] autonomously. Here's the thing. If they can beat all these challenges, yeah,
[01:20:47] can beat all these challenges, yeah, robotics would be really cool. I don't
[01:20:48] robotics would be really cool. I don't know how long that's that's one I'd
[01:20:50] know how long that's that's one I'd actually be willing to believe in a
[01:20:53] actually be willing to believe in a couple decades.
[01:20:54] couple decades. >> Have you seen them ch them Chinese
[01:20:55] >> Have you seen them ch them Chinese robots? I know you've seen them. the
[01:20:56] robots? I know you've seen them. the uni, what's it called? The one that can
[01:20:58] uni, what's it called? The one that can dance and that, but they can't really do
[01:21:00] dance and that, but they can't really do human things.
[01:21:01] human things. >> Well, it's just it is pretty
[01:21:03] >> Well, it's just it is pretty mindblowing.
[01:21:04] mindblowing. >> Robotics are cool. I like I'm
[01:21:06] >> Robotics are cool. I like I'm not going to pretend. I don't think
[01:21:07] not going to pretend. I don't think robots are cool. I wish they were
[01:21:09] robots are cool. I wish they were building robots and actually doing cool
[01:21:11] building robots and actually doing cool I wish the tech industry still
[01:21:13] I wish the tech industry still made fun stuff and interesting stuff.
[01:21:15] made fun stuff and interesting stuff. Instead, we get these large
[01:21:16] Instead, we get these large language models. But with AI plus
[01:21:19] language models. But with AI plus robotics is, you know, I was in San
[01:21:21] robotics is, you know, I was in San Francisco and I went to this massive um
[01:21:22] Francisco and I went to this massive um incubator there. And when I'd gone there
[01:21:25] incubator there. And when I'd gone there three years earlier, it was all software
[01:21:26] three years earlier, it was all software startups, right? And when I went back
[01:21:28] startups, right? And when I went back three years later, it was all these
[01:21:30] three years later, it was all these robot startups. And I remember saying to
[01:21:31] robot startups. And I remember saying to the founder of the incubator, I was
[01:21:32] the founder of the incubator, I was like, "Why is everything robots now?"
[01:21:34] like, "Why is everything robots now?" There was this one robot where it was
[01:21:36] There was this one robot where it was just the arm and it had a frying pan on
[01:21:38] just the arm and it had a frying pan on it. Yeah.
[01:21:38] it. Yeah. >> And it whole thing is it cooks for you.
[01:21:40] >> And it whole thing is it cooks for you. >> Yeah.
[01:21:40] >> Yeah. >> So it was he was showing me it cooking
[01:21:42] >> So it was he was showing me it cooking whatever. And he goes, "Well, you know
[01:21:43] whatever. And he goes, "Well, you know the arm." He goes, "The the hardware
[01:21:45] the arm." He goes, "The the hardware part, the physical parts,
[01:21:47] part, the physical parts, >> that's always been fairly cheap." Yeah.
[01:21:49] >> that's always been fairly cheap." Yeah. >> He goes, "The expensive part was the
[01:21:51] >> He goes, "The expensive part was the intelligence. And now that's come down
[01:21:52] intelligence. And now that's come down to pennies." So what you're seeing is
[01:21:54] to pennies." So what you're seeing is this explosion in the robotics industry
[01:21:56] this explosion in the robotics industry because robotics is a function of
[01:21:57] because robotics is a function of intelligence plus hardware. We've always
[01:21:59] intelligence plus hardware. We've always had the
[01:21:59] had the >> and a ton of data though as well and the
[01:22:01] >> and a ton of data though as well and the data is very expensive.
[01:22:03] data is very expensive. >> Yeah.
[01:22:03] >> Yeah. >> The thing is cyber cabs rolled out real
[01:22:06] >> The thing is cyber cabs rolled out real slow. It's going to take a long time. It
[01:22:08] slow. It's going to take a long time. It could be a threat if they do a robot
[01:22:10] could be a threat if they do a robot that could replace a human job. Sure it
[01:22:12] that could replace a human job. Sure it could. But that human jobs are
[01:22:14] could. But that human jobs are multifaceted. Human jobs change with
[01:22:16] multifaceted. Human jobs change with environments. And also a lot of human
[01:22:18] environments. And also a lot of human jobs that you might think of like I
[01:22:20] jobs that you might think of like I don't know dishwashing robot for
[01:22:21] don't know dishwashing robot for example.
[01:22:22] example. >> Yeah.
[01:22:23] >> Yeah. some guy at a restaurant isn't paying 10
[01:22:25] some guy at a restaurant isn't paying 10 20 grand for a robot to replace the job
[01:22:27] 20 grand for a robot to replace the job that they're already not paying enough
[01:22:28] that they're already not paying enough for. The point is, yeah, it could if you
[01:22:31] for. The point is, yeah, it could if you can replace the jobs. That is not what
[01:22:34] can replace the jobs. That is not what we're talking about with this.
[01:22:35] we're talking about with this. >> Yeah. I I just I just I ask these
[01:22:37] >> Yeah. I I just I just I ask these questions not because I'm trying to be
[01:22:38] questions not because I'm trying to be like I actually I'm trying to form my
[01:22:40] like I actually I'm trying to form my own opinion on these things and
[01:22:42] own opinion on these things and >> I I do think, you know, as it's written
[01:22:45] >> I I do think, you know, as it's written there, it says AI will replace all human
[01:22:48] there, it says AI will replace all human jobs. Obviously not. Obviously, that's
[01:22:50] jobs. Obviously not. Obviously, that's Yeah.
[01:22:50] Yeah. >> But um I'm trying to figure out if the
[01:22:52] >> But um I'm trying to figure out if the truth is somewhere in the middle that
[01:22:53] truth is somewhere in the middle that there's a certain type of job which
[01:22:55] there's a certain type of job which actually humans probably shouldn't have
[01:22:57] actually humans probably shouldn't have ever been doing really.
[01:22:59] ever been doing really. >> Um if you think back through history,
[01:23:00] >> Um if you think back through history, there was someone's job just to sit in
[01:23:01] there was someone's job just to sit in an elevator and press the buttons.
[01:23:03] an elevator and press the buttons. >> That's an example of a job that humans
[01:23:04] >> That's an example of a job that humans probably shouldn't have been doing. And
[01:23:06] probably shouldn't have been doing. And as technology gets more advanced, it
[01:23:07] as technology gets more advanced, it takes on a lot of that
[01:23:09] takes on a lot of that >> sort of automated monotonous stuff.
[01:23:12] >> sort of automated monotonous stuff. >> Right? The thing is with this particular
[01:23:14] >> Right? The thing is with this particular thing that I know that this is from,
[01:23:16] thing that I know that this is from, it's a specific blog I wrote. I was
[01:23:17] it's a specific blog I wrote. I was explicitly talking about generative AI
[01:23:19] explicitly talking about generative AI though. I was explicitly [clears throat]
[01:23:20] though. I was explicitly [clears throat] talking about people when they say this
[01:23:22] talking about people when they say this they are referring to that.
[01:23:23] they are referring to that. >> So you're not talking about agentic AI
[01:23:25] >> So you're not talking about agentic AI which is
[01:23:25] which is >> agentic AI is LLMs. Agentic AI is just a
[01:23:28] >> agentic AI is LLMs. Agentic AI is just a fancy way of saying an LLM talking to
[01:23:29] fancy way of saying an LLM talking to another LLM with a harness on top. That
[01:23:31] another LLM with a harness on top. That is still LLM. Agentic AI is one of the
[01:23:34] is still LLM. Agentic AI is one of the big the bigger lies they to tell. It's
[01:23:36] big the bigger lies they to tell. It's like when you hear agent you're meant to
[01:23:37] like when you hear agent you're meant to think autonomous AI can do what you
[01:23:39] think autonomous AI can do what you want. It's still LLMs. It's still LM
[01:23:41] want. It's still LLMs. It's still LM talking to other LMLs
[01:23:42] talking to other LMLs >> taking screenshots and putting them in
[01:23:44] >> taking screenshots and putting them in LLM and stuff.
[01:23:45] LLM and stuff. >> Oh god. Yeah.
[01:23:46] >> Oh god. Yeah. >> Okay. But but you know I could I could
[01:23:47] >> Okay. But but you know I could I could make the case that
[01:23:50] make the case that I'm just thinking about my personal
[01:23:51] I'm just thinking about my personal usage. I definitely use agents to do
[01:23:54] usage. I definitely use agents to do things that I would have previously
[01:23:55] things that I would have previously asked people to do. It's not to say that
[01:23:56] asked people to do. It's not to say that I didn't I still don't hire cuz we're
[01:23:58] I didn't I still don't hire cuz we're hiring like crazy.
[01:23:59] hiring like crazy. >> Yeah.
[01:23:59] >> Yeah. >> And I still in that particular function.
[01:24:01] >> And I still in that particular function. I'm thinking about like the chief of
[01:24:02] I'm thinking about like the chief of staff role. So my chief of staff would
[01:24:05] staff role. So my chief of staff would have triaged all of my inboxes
[01:24:06] have triaged all of my inboxes previously and put them somewhere and
[01:24:08] previously and put them somewhere and told me about them or maybe once upon a
[01:24:10] told me about them or maybe once upon a time shown me a piece of paper back in
[01:24:11] time shown me a piece of paper back in the day. I guess now my chief of staff
[01:24:13] the day. I guess now my chief of staff is no longer doing that job. You still
[01:24:15] is no longer doing that job. You still have a chief of staff though.
[01:24:16] have a chief of staff though. >> This is what I'm saying. They're doing
[01:24:17] >> This is what I'm saying. They're doing other things,
[01:24:18] other things, >> right? But the thing is again what you
[01:24:20] >> right? But the thing is again what you were describing is
[01:24:22] were describing is fairly basic automation. I don't know
[01:24:24] fairly basic automation. I don't know what the tasks are triaging.
[01:24:26] what the tasks are triaging. >> Basic spend a trillion dollars on
[01:24:28] >> Basic spend a trillion dollars on triaging email. Like that's the the
[01:24:29] triaging email. Like that's the the promise. If they'd spent $10 billion and
[01:24:32] promise. If they'd spent $10 billion and this was much smaller and you I go cool
[01:24:34] this was much smaller and you I go cool software. Yay. A lot of the things that
[01:24:35] software. Yay. A lot of the things that people are impressed with like script
[01:24:37] people are impressed with like script stuff as well. It's just LM's doing
[01:24:38] stuff as well. It's just LM's doing Python. You should be impressed by
[01:24:40] Python. You should be impressed by Python code. Python's incredible. You
[01:24:42] Python code. Python's incredible. You can scrape websites. You can download
[01:24:43] can scrape websites. You can download It's awesome. But the point I'm
[01:24:46] It's awesome. But the point I'm making is none of this would be anywhere
[01:24:48] making is none of this would be anywhere near as much of a problem if they didn't
[01:24:50] near as much of a problem if they didn't ask for all of the attention, all of the
[01:24:52] ask for all of the attention, all of the money, and promise the world. It's their
[01:24:54] money, and promise the world. It's their promises that are the problem. And the
[01:24:55] promises that are the problem. And the journalists who went along with it, and
[01:24:56] journalists who went along with it, and the analysts and the Twitter people who
[01:24:58] the analysts and the Twitter people who went along with this, saying that this
[01:24:59] went along with this, saying that this would change everything and replace
[01:25:01] would change everything and replace everything and leaving the realm of
[01:25:03] everything and leaving the realm of reality. Is there any technological
[01:25:05] reality. Is there any technological innovation through history that was
[01:25:07] innovation through history that was really, really game-changing where that
[01:25:08] really, really game-changing where that didn't happen?
[01:25:11] didn't happen? I mean
[01:25:13] I mean the internet
[01:25:14] the internet >> I mean people overpromised that
[01:25:15] >> I mean people overpromised that >> I mean they overpromised on the
[01:25:17] >> I mean they overpromised on the businesses but I've read through a great
[01:25:19] businesses but I've read through a great many pieces about the early internet a
[01:25:21] many pieces about the early internet a lot of people were excited but hesitant
[01:25:25] lot of people were excited but hesitant they were worried that there was not
[01:25:26] they were worried that there was not enough demand but they were still like
[01:25:28] enough demand but they were still like oh yeah this could have potential
[01:25:30] oh yeah this could have potential ramifications if it happened. People
[01:25:32] ramifications if it happened. People were not super negative about the
[01:25:35] were not super negative about the internet. A lot of the skeptics were
[01:25:36] internet. A lot of the skeptics were saying we're worried about an overload
[01:25:37] saying we're worried about an overload of bad information. Look at where we
[01:25:39] of bad information. Look at where we are. A lot of people were worried about
[01:25:41] are. A lot of people were worried about the social consequences of everyone
[01:25:43] the social consequences of everyone talking online, which they were correct
[01:25:44] talking online, which they were correct about. With the economic things, they
[01:25:46] about. With the economic things, they were specifically talking about like the
[01:25:48] were specifically talking about like the globe, which I think made hundreds of
[01:25:49] globe, which I think made hundreds of thousands of dollars and had like a I
[01:25:52] thousands of dollars and had like a I think a billion dollar market cap, but
[01:25:53] think a billion dollar market cap, but they were talking.
[01:25:54] they were talking. >> Yeah, there was massive hype in the com
[01:25:56] >> Yeah, there was massive hype in the com era.
[01:25:56] era. >> I read a lot of those stories. The hype
[01:25:58] >> I read a lot of those stories. The hype was nowhere in it. You didn't have
[01:25:59] was nowhere in it. You didn't have articles everywhere that were saying if
[01:26:02] articles everywhere that were saying if you don't get online, you'll be left
[01:26:03] you don't get online, you'll be left behind. You didn't have professional
[01:26:05] behind. You didn't have professional consequences. Nick Sesh mentioned his
[01:26:07] consequences. Nick Sesh mentioned his blog earlier. He described this thing
[01:26:09] blog earlier. He described this thing global uh AI sisterating global
[01:26:11] global uh AI sisterating global decision-m where he said that you have
[01:26:14] decision-m where he said that you have businesses you work at where if you
[01:26:16] businesses you work at where if you don't say that you're more productive
[01:26:18] don't say that you're more productive with AI whether or not it's true is
[01:26:20] with AI whether or not it's true is irrelevant you have professional
[01:26:22] irrelevant you have professional consequences you can get fired there are
[01:26:24] consequences you can get fired there are people having to AI wash their jobs by
[01:26:26] people having to AI wash their jobs by saying AI did it otherwise their bosses
[01:26:28] saying AI did it otherwise their bosses who don't do will get mad at them
[01:26:31] who don't do will get mad at them this did not happen with the internet it
[01:26:34] this did not happen with the internet it was not present and part of the thing is
[01:26:36] was not present and part of the thing is social media was not like it is today
[01:26:38] social media was not like it is today the kind of uh was it decentralization
[01:26:40] the kind of uh was it decentralization of media in general has caused this as
[01:26:43] of media in general has caused this as well and also the fact of day trading
[01:26:45] well and also the fact of day trading there's so many different things that
[01:26:46] there's so many different things that are different it's crazy
[01:26:48] are different it's crazy >> I I do think AI is different from the
[01:26:50] >> I I do think AI is different from the internet in part if you just measured it
[01:26:52] internet in part if you just measured it on the speed of adoption especially if
[01:26:54] on the speed of adoption especially if we just think about generative AI AI
[01:26:56] we just think about generative AI AI >> but the this adoption of the internet
[01:26:58] >> but the this adoption of the internet required physical connections to your
[01:27:00] required physical connections to your house the adoption of generative AI
[01:27:02] house the adoption of generative AI involves having a web browser it took a
[01:27:04] involves having a web browser it took a vast amount of effort to bring internet
[01:27:06] vast amount of effort to bring internet to people Even with dialup connections,
[01:27:08] to people Even with dialup connections, it still required the distribution
[01:27:10] it still required the distribution >> and that's why it was so slow and there
[01:27:12] >> and that's why it was so slow and there was less, you know, there was less hype
[01:27:13] was less, you know, there was less hype than AI. I do agree that there's way
[01:27:15] than AI. I do agree that there's way more hype and we again going back to
[01:27:17] more hype and we again going back to this point that we're clustering AI in
[01:27:20] this point that we're clustering AI in this big category of lots of different
[01:27:21] this big category of lots of different things.
[01:27:22] things. >> There's generative AI.
[01:27:23] >> There's generative AI. >> There's generative AI. There's like real
[01:27:24] >> There's generative AI. There's like real world AI.
[01:27:25] world AI. >> Generative AI is explicitly what I'm
[01:27:26] >> Generative AI is explicitly what I'm talking about here. When bosses are
[01:27:28] talking about here. When bosses are saying you need to use AI, they're not
[01:27:29] saying you need to use AI, they're not saying I need you to go and buy a
[01:27:30] saying I need you to go and buy a Unibeam robot. They're saying use LLM so
[01:27:33] Unibeam robot. They're saying use LLM so that I and that's the thing. They have
[01:27:35] that I and that's the thing. They have this theory, the era of the business
[01:27:37] this theory, the era of the business idiot where it's like we are ruled by
[01:27:38] idiot where it's like we are ruled by people that don't do work because nobody
[01:27:40] people that don't do work because nobody who actually does a bunch of work who
[01:27:42] who actually does a bunch of work who really is productive is harassing
[01:27:44] really is productive is harassing someone who works for them for not being
[01:27:46] someone who works for them for not being productive enough.
[01:27:47] productive enough. >> They're not they don't have the time.
[01:27:49] >> They're not they don't have the time. They're doing work. Someone who is
[01:27:50] They're doing work. Someone who is sitting there with the ingratiation
[01:27:51] sitting there with the ingratiation machine that's telling them that every
[01:27:53] machine that's telling them that every beautiful idea out of their messy little
[01:27:55] beautiful idea out of their messy little skull is amazing. Yeah. They're going,
[01:27:57] skull is amazing. Yeah. They're going, "Damn, this thing says I'm a genius. Why
[01:27:59] "Damn, this thing says I'm a genius. Why are you not using the genius machine to
[01:28:00] are you not using the genius machine to do more work?" And yeah, if you're a
[01:28:03] do more work?" And yeah, if you're a boss that goes to lunch, leaves lunch,
[01:28:04] boss that goes to lunch, leaves lunch, and sometimes reads your emails, LM are
[01:28:06] and sometimes reads your emails, LM are magic.
[01:28:07] magic. >> I, you know, one of the most compelling
[01:28:08] >> I, you know, one of the most compelling arguments I have for the overhype of AI
[01:28:13] arguments I have for the overhype of AI >> in a world where everybody has access to
[01:28:14] >> in a world where everybody has access to these tools, whatever the
[01:28:16] these tools, whatever the [clears throat] tools can do, would
[01:28:18] [clears throat] tools can do, would largely be commoditized. What the tools
[01:28:20] largely be commoditized. What the tools can't do, which one could say is the
[01:28:23] can't do, which one could say is the human taste, judgment, you could say
[01:28:25] human taste, judgment, you could say it's people, skills, whatever you want
[01:28:27] it's people, skills, whatever you want to say, is now going to be the valuable
[01:28:29] to say, is now going to be the valuable thing because the scarce and the hard
[01:28:31] thing because the scarce and the hard becomes the most valuable through
[01:28:33] becomes the most valuable through history and the commoditized becomes the
[01:28:36] history and the commoditized becomes the least valuable. So the very nature that
[01:28:38] least valuable. So the very nature that we're commoditizing, the generation of
[01:28:40] we're commoditizing, the generation of content or whatever you want to call it,
[01:28:41] content or whatever you want to call it, code means that's actually not where the
[01:28:43] code means that's actually not where the value will acrue as for the user. And
[01:28:46] value will acrue as for the user. And actually if you think about what it
[01:28:48] actually if you think about what it takes to now make something that is
[01:28:50] takes to now make something that is objectively great if an AI can do it
[01:28:55] objectively great if an AI can do it then it's not the the great thing is not
[01:28:57] then it's not the the great thing is not of value.
[01:28:58] of value. >> So so I think a lot I've been thinking a
[01:28:59] >> So so I think a lot I've been thinking a lot actually about how
[01:29:01] lot actually about how >> how do you um avoid the temptation of
[01:29:05] >> how do you um avoid the temptation of sloppification of the things you make
[01:29:06] sloppification of the things you make the value you put into the world. It's
[01:29:08] the value you put into the world. It's very simple example that people will be
[01:29:10] very simple example that people will be able to relate to. If you use chat GBT
[01:29:12] able to relate to. If you use chat GBT or anthropic, you know, Claude to make
[01:29:14] or anthropic, you know, Claude to make your LinkedIn posts, let's say,
[01:29:16] your LinkedIn posts, let's say, >> they will be LinkedIn posts because
[01:29:18] >> they will be LinkedIn posts because everybody else is using them. And
[01:29:19] everybody else is using them. And actually, a great LinkedIn post now is
[01:29:22] actually, a great LinkedIn post now is someone who doesn't use them and makes
[01:29:23] someone who doesn't use them and makes something that's like irreplaceably
[01:29:25] something that's like irreplaceably human,
[01:29:25] human, >> right?
[01:29:26] >> right? >> And deeper and more personal N of one
[01:29:31] >> And deeper and more personal N of one lived experience.
[01:29:32] lived experience. >> Yeah.
[01:29:32] >> Yeah. >> All these things that AI can't do. And I
[01:29:34] >> All these things that AI can't do. And I think that's a compelling argument that
[01:29:36] think that's a compelling argument that actually the commodity tools produce
[01:29:38] actually the commodity tools produce commodity outcomes. So everyone has
[01:29:40] commodity outcomes. So everyone has access to these things and what's
[01:29:41] access to these things and what's changed? Like really like what
[01:29:43] changed? Like really like what >> the slopification we've we've got a
[01:29:45] >> the slopification we've we've got a bunch of slop but these people were
[01:29:46] bunch of slop but these people were halfassing their jobs before. It's just
[01:29:48] halfassing their jobs before. It's just a halfass arcery machine and it's just
[01:29:50] a halfass arcery machine and it's just it's it's the thing. It's what I'm
[01:29:52] it's it's the thing. It's what I'm talking about with the slot blogs. It's
[01:29:53] talking about with the slot blogs. It's like it's it yeah people that gave you
[01:29:55] like it's it yeah people that gave you dog before have now got the dog
[01:29:57] dog before have now got the dog machine to pump out dog It's
[01:29:59] machine to pump out dog It's so there's a guy called Carl Brown uh
[01:30:01] so there's a guy called Carl Brown uh internet bucks. Awesome guy. Great
[01:30:03] internet bucks. Awesome guy. Great software engineer. He he said I might
[01:30:05] software engineer. He he said I might have said this earlier. So, it makes the
[01:30:06] have said this earlier. So, it makes the easy things easy, the hard things
[01:30:08] easy things easy, the hard things harder. When you know you're doing a
[01:30:09] harder. When you know you're doing a really distinct small script for
[01:30:10] really distinct small script for something and it can plop that out. It's
[01:30:12] something and it can plop that out. It's awesome. I used Claude the other day for
[01:30:14] awesome. I used Claude the other day for something useful. My kid loves
[01:30:16] something useful. My kid loves Minecraft. I was trying to fix a
[01:30:18] Minecraft. I was trying to fix a broken mod cuz he loves his wither
[01:30:19] broken mod cuz he loves his wither storm. It's awesome.
[01:30:21] storm. It's awesome. >> And it still took me half an hour and
[01:30:23] >> And it still took me half an hour and kept getting things wrong. What do you
[01:30:24] kept getting things wrong. What do you use AI for? Generative.
[01:30:26] use AI for? Generative. >> I really don't. I don't use it
[01:30:27] >> I really don't. I don't use it >> with Bloomberg terminal. I use AskB,
[01:30:29] >> with Bloomberg terminal. I use AskB, which is just when it's like requesting
[01:30:31] which is just when it's like requesting the consensus analyst estimates for
[01:30:33] the consensus analyst estimates for Nvidia,
[01:30:33] Nvidia, >> but otherwise you don't use it.
[01:30:35] >> but otherwise you don't use it. >> No. So, how do you know it's bad? I've
[01:30:37] >> No. So, how do you know it's bad? I've used it. I've put it through its paces.
[01:30:38] used it. I've put it through its paces. I've used it to try and do financial
[01:30:40] I've used it to try and do financial models and found one error and
[01:30:41] models and found one error and immediately be like, "Ah, I've never
[01:30:43] immediately be like, "Ah, I've never been particularly impressed." The one
[01:30:45] been particularly impressed." The one thing I will defend it on is it's really
[01:30:46] thing I will defend it on is it's really good for like tech support. Like I have
[01:30:48] good for like tech support. Like I have this thing called Synergy in my New York
[01:30:50] this thing called Synergy in my New York New York place I go to. I have this
[01:30:52] New York place I go to. I have this monitor where I have a MacBook and a PC
[01:30:54] monitor where I have a MacBook and a PC laptop and this thing Synergy for using
[01:30:56] laptop and this thing Synergy for using the same mouse and keyboard.
[01:30:57] the same mouse and keyboard. >> Dropping a giant
[01:31:00] >> Dropping a giant troubleshooting log into this thing and
[01:31:02] troubleshooting log into this thing and going, "What's wrong?" And it going,
[01:31:03] going, "What's wrong?" And it going, "This is wrong." Yeah, super useful. Is
[01:31:05] "This is wrong." Yeah, super useful. Is that trillion dollars? No. Is that a $2
[01:31:07] that trillion dollars? No. Is that a $2 trillion company? No. Pretty use.
[01:31:09] trillion company? No. Pretty use. >> Better than Google though, right? Better
[01:31:10] >> Better than Google though, right? Better than Google search.
[01:31:11] than Google search. >> I know. I mean, yeah. Remember,
[01:31:13] >> I know. I mean, yeah. Remember, >> do you use Google search still?
[01:31:14] >> do you use Google search still? >> I try. I have to push the crap
[01:31:16] >> I try. I have to push the crap out of the way. And
[01:31:18] out of the way. And >> I can't remember the last time I did a
[01:31:20] >> I can't remember the last time I did a Google search.
[01:31:21] Google search. >> Christ, I find myself using Bing
[01:31:22] >> Christ, I find myself using Bing sometimes. I know. I hate saying it,
[01:31:24] sometimes. I know. I hate saying it, too. But I have to scroll past the AI
[01:31:27] too. But I have to scroll past the AI crap cuz I want the good stuff. I want
[01:31:29] crap cuz I want the good stuff. I want the I want the actual links to stuff so
[01:31:31] the I want the actual links to stuff so that I can read the thing and go. But
[01:31:33] that I can read the thing and go. But you can ask the AI to give you the
[01:31:35] you can ask the AI to give you the links.
[01:31:36] links. >> Yeah. And it doesn't do a particularly
[01:31:38] >> Yeah. And it doesn't do a particularly good job. Like my
[01:31:39] good job. Like my >> So say that the other day my iPad wasn't
[01:31:41] >> So say that the other day my iPad wasn't turning on and it was doing this funny
[01:31:42] turning on and it was doing this funny little thing on the screen. You think
[01:31:44] little thing on the screen. You think that it's better to type that into
[01:31:46] that it's better to type that into Google than
[01:31:46] Google than >> Oh, no. I must be clear that may be the
[01:31:48] >> Oh, no. I must be clear that may be the only LLM use case I defend. The
[01:31:50] only LLM use case I defend. The troubleshooting thing is awesome for it.
[01:31:52] troubleshooting thing is awesome for it. I It's the the one weakness I have. It's
[01:31:54] I It's the the one weakness I have. It's like genuinely being able to drop a log
[01:31:56] like genuinely being able to drop a log into it. That's awesome. Again, that is
[01:31:59] into it. That's awesome. Again, that is not what they're selling it as. They're
[01:32:01] not what they're selling it as. They're not selling it as a useful little tool.
[01:32:03] not selling it as a useful little tool. They're selling it as the uh software as
[01:32:06] They're selling it as the uh software as the thing that will change everything
[01:32:07] the thing that will change everything that will replace all jobs that will do
[01:32:09] that will replace all jobs that will do this and that. It's not like they sold
[01:32:11] this and that. It's not like they sold it as a quirky bit of software.
[01:32:13] it as a quirky bit of software. >> No, you are right. They are, you know,
[01:32:14] >> No, you are right. They are, you know, telling us that it is going to replace
[01:32:16] telling us that it is going to replace everything. But funnily enough, the
[01:32:17] everything. But funnily enough, the critics are saying that as well.
[01:32:19] critics are saying that as well. >> Which one I mean I mean
[01:32:20] >> Which one I mean I mean >> they are like the Jeffrey Hintons of the
[01:32:22] >> they are like the Jeffrey Hintons of the world. you know, even people that have
[01:32:23] world. you know, even people that have left the safety team in chat who who
[01:32:26] left the safety team in chat who who I've sat here with the these are critics
[01:32:28] I've sat here with the these are critics that are that are warning of the impacts
[01:32:30] that are that are warning of the impacts it's going to have on the world. It's
[01:32:32] it's going to have on the world. It's weird how all these critics also have
[01:32:34] weird how all these critics also have vested interest in AI doing well though.
[01:32:36] vested interest in AI doing well though. Daniel, former open AI guy, AI 2027
[01:32:39] Daniel, former open AI guy, AI 2027 written with the Star Codeex guy that
[01:32:40] written with the Star Codeex guy that was nothing more than badly written
[01:32:42] was nothing more than badly written science fiction that he's already had to
[01:32:44] science fiction that he's already had to walk back.
[01:32:44] walk back. >> You know, he could have made more money
[01:32:45] >> You know, he could have made more money by staying at chat.
[01:32:47] by staying at chat. >> Could he?
[01:32:48] >> Could he? >> I mean, looks like he lost
[01:32:50] >> I mean, looks like he lost >> if he had options early. it sticking
[01:32:52] >> if he had options early. it sticking around.
[01:32:53] around. >> Did he lose the options? How much do
[01:32:54] >> Did he lose the options? How much do they
[01:32:55] they >> You're not saying that they're they're
[01:32:56] >> You're not saying that they're they're being critical. They're not critical of
[01:32:59] being critical. They're not critical of the companies themselves. They're not
[01:33:00] the companies themselves. They're not critical of the stealing. They're not
[01:33:01] critical of the stealing. They're not critical of the environmental damage.
[01:33:03] critical of the environmental damage. They're not critical of the fact that
[01:33:05] They're not critical of the fact that you cannot rely on the answers. They're
[01:33:06] you cannot rely on the answers. They're critical of this big scary boogeyman out
[01:33:09] critical of this big scary boogeyman out in the future where it's like, "Oh, I'm
[01:33:12] in the future where it's like, "Oh, I'm scared of when this becomes so powerful
[01:33:14] scared of when this becomes so powerful and everyone should talk to me about how
[01:33:15] and everyone should talk to me about how scary and powerful it is." They're not
[01:33:17] scary and powerful it is." They're not saying, "Hey, here are the harms today.
[01:33:19] saying, "Hey, here are the harms today. Here are the things we're actually
[01:33:20] Here are the things we're actually looking at today. Here are the social
[01:33:22] looking at today. Here are the social problems of having this automated way of
[01:33:25] problems of having this automated way of spewing out slop, of filling our feeds
[01:33:28] spewing out slop, of filling our feeds with crap, of having information that
[01:33:30] with crap, of having information that will pop up that is presented even with
[01:33:32] will pop up that is presented even with the little disclaimer thing of saying,
[01:33:33] the little disclaimer thing of saying, "Yeah, sometimes this gets wrong."
[01:33:35] "Yeah, sometimes this gets wrong." So, in the tiniest words possible, they
[01:33:37] So, in the tiniest words possible, they don't talk about the fact that these
[01:33:39] don't talk about the fact that these things are trained on stealing millions
[01:33:41] things are trained on stealing millions of people's work. But on that last point
[01:33:43] of people's work. But on that last point where you say that it's going to get
[01:33:44] where you say that it's going to get progressively more intelligent and when
[01:33:45] progressively more intelligent and when it does, it will be a danger.
[01:33:47] it does, it will be a danger. >> Yeah. Would you agree with the statement
[01:33:50] >> Yeah. Would you agree with the statement that artificial intelligence has gotten
[01:33:52] that artificial intelligence has gotten more intelligent
[01:33:54] more intelligent if you measure it based on any sort of
[01:33:56] if you measure it based on any sort of measure of intelligence one might use?
[01:33:58] measure of intelligence one might use? >> It's got better on the tests that are
[01:33:59] >> It's got better on the tests that are rigged for the models. It's got better
[01:34:01] rigged for the models. It's got better at tests where you can train for the
[01:34:03] at tests where you can train for the test.
[01:34:03] test. >> Okay, so it's got better at
[01:34:04] >> Okay, so it's got better at >> it's got better at tests that they're
[01:34:06] >> it's got better at tests that they're intentionally trained for.
[01:34:08] intentionally trained for. >> So if you logged the rate of improvement
[01:34:10] >> So if you logged the rate of improvement on a graph, it would look something like
[01:34:13] on a graph, it would look something like this,
[01:34:14] this, >> right?
[01:34:14] >> right? >> You agree? in terms of what it's capable
[01:34:16] >> You agree? in terms of what it's capable of doing.
[01:34:18] of doing. There we go. Yeah,
[01:34:19] There we go. Yeah, >> cuz it's not it's not got new features.
[01:34:21] >> cuz it's not it's not got new features. You'll notice that outside of OpenAI and
[01:34:23] You'll notice that outside of OpenAI and Anthropic the VA when you remove the
[01:34:26] Anthropic the VA when you remove the coding startups, there's basically no
[01:34:28] coding startups, there's basically no successful AI startup company.
[01:34:30] successful AI startup company. >> So, we agree that it's got better. It's
[01:34:32] >> So, we agree that it's got better. It's got more capable
[01:34:34] got more capable at doing things.
[01:34:35] at doing things. >> Yeah. Okay. Over time, AI's got more
[01:34:38] >> Yeah. Okay. Over time, AI's got more capable. If we imagine that trajectory
[01:34:41] capable. If we imagine that trajectory will continue, it will get more capable.
[01:34:44] will continue, it will get more capable. Then at some point it does cross you
[01:34:46] Then at some point it does cross you know this is what they say to me it
[01:34:48] know this is what they say to me it crosses human intelligence and at such
[01:34:50] crosses human intelligence and at such time
[01:34:51] time >> will it not start to do some of the jobs
[01:34:54] >> will it not start to do some of the jobs that people are doing today
[01:34:56] that people are doing today >> outside of software engineering remove
[01:34:57] >> outside of software engineering remove software because I will concede software
[01:34:58] software because I will concede software engineering it's got better at that
[01:35:00] engineering it's got better at that outside of software engineering where
[01:35:02] outside of software engineering where >> so the chief of staff things that admin
[01:35:04] >> so the chief of staff things that admin >> okay so it's got better admin video
[01:35:06] >> okay so it's got better admin video generation photo generation
[01:35:08] generation photo generation >> text generation theoretically coding
[01:35:12] >> text generation theoretically coding >> right
[01:35:12] >> right >> and then I'd say agentic workflows. So
[01:35:15] >> and then I'd say agentic workflows. So >> what is an agentic workflow?
[01:35:16] >> what is an agentic workflow? >> So automated workflows where you're
[01:35:17] >> So automated workflows where you're doing the same I mean a good example is
[01:35:20] doing the same I mean a good example is looking at the backend data of the dire
[01:35:22] looking at the backend data of the dire of a CEO
[01:35:22] of a CEO >> summarizing
[01:35:23] >> summarizing >> looking at all of the data ingesting all
[01:35:24] >> looking at all of the data ingesting all of it going out into the internet and
[01:35:26] of it going out into the internet and searching who Ed is
[01:35:27] searching who Ed is >> looking at every interview you've ever
[01:35:29] >> looking at every interview you've ever done ever.
[01:35:29] done ever. >> Uhhuh.
[01:35:30] >> Uhhuh. >> This is summarizing and generating
[01:35:32] >> This is summarizing and generating >> making a little model on you know the
[01:35:34] >> making a little model on you know the things people want to know from Ed.
[01:35:36] things people want to know from Ed. >> Producing a report sending that to my
[01:35:37] >> Producing a report sending that to my inbox.
[01:35:38] inbox. >> Me getting a 20 30 40 50page report on
[01:35:40] >> Me getting a 20 30 40 50page report on Ed before he arrives.
[01:35:42] Ed before he arrives. >> This is all basically the same thing. I
[01:35:43] >> This is all basically the same thing. I think it's been doing for years though.
[01:35:44] think it's been doing for years though. It's It's not really new capabilities.
[01:35:46] It's It's not really new capabilities. >> Research. It's It's
[01:35:48] >> Research. It's It's >> still the same things. They've had web
[01:35:50] >> still the same things. They've had web search for years. They've had report
[01:35:51] search for years. They've had report generation for years.
[01:35:52] generation for years. >> Well, we couldn't generate
[01:35:55] >> Well, we couldn't generate highquality videos that are like
[01:35:56] highquality videos that are like indistinguishable from cameras. Seed
[01:35:59] indistinguishable from cameras. Seed dance and these ones that look like
[01:36:00] dance and these ones that look like movies.
[01:36:01] movies. >> I mean, they
[01:36:02] >> I mean, they >> are incredible.
[01:36:03] >> are incredible. >> So, I'm saying the point I'm trying to
[01:36:04] >> So, I'm saying the point I'm trying to make is that if we imagine that over the
[01:36:06] make is that if we imagine that over the last 10 years there has been a rate of
[01:36:07] last 10 years there has been a rate of improvement in terms of capabilities and
[01:36:09] improvement in terms of capabilities and output and quality. We've seen
[01:36:10] output and quality. We've seen hallucinations drop. We've seen the
[01:36:12] hallucinations drop. We've seen the models get more quote unquote
[01:36:14] models get more quote unquote intelligent, get better at, you know, if
[01:36:16] intelligent, get better at, you know, if you did give it an IQ test, it's getting
[01:36:17] you did give it an IQ test, it's getting higher scores than it was 10 years ago.
[01:36:19] higher scores than it was 10 years ago. We agree that there's been a upward
[01:36:21] We agree that there's been a upward motion of improvement.
[01:36:22] motion of improvement. >> This is pretty much how machine learning
[01:36:23] >> This is pretty much how machine learning goes when you feed it more data.
[01:36:24] goes when you feed it more data. >> Exactly. And you put more compute behind
[01:36:26] >> Exactly. And you put more compute behind it. So if this continues,
[01:36:30] it. So if this continues, what does the future look like? So the
[01:36:32] what does the future look like? So the rebuttal I was expecting to hear is that
[01:36:34] rebuttal I was expecting to hear is that it won't continue. And actually,
[01:36:35] it won't continue. And actually, >> I actually don't think it I think that
[01:36:37] >> I actually don't think it I think that there are hard limits that we're going
[01:36:39] there are hard limits that we're going to hit. So you do believe in that
[01:36:40] to hit. So you do believe in that there's a hard limit somewhere.
[01:36:42] there's a hard limit somewhere. >> We've kind of already hit the
[01:36:43] >> We've kind of already hit the diminishing returns level because for
[01:36:45] diminishing returns level because for example video generation which is by the
[01:36:48] example video generation which is by the way far less an American concern
[01:36:50] way far less an American concern anymore. OpenAI shut down Sora. I think
[01:36:53] anymore. OpenAI shut down Sora. I think you can still use the API but
[01:36:54] you can still use the API but nevertheless look at the look around you
[01:36:56] nevertheless look at the look around you with the amount of stuff in the crew you
[01:36:58] with the amount of stuff in the crew you need to get a shot. People think the
[01:36:59] need to get a shot. People think the movies are just shot by shot by shot and
[01:37:01] movies are just shot by shot by shot and they just magically happen. When you've
[01:37:02] they just magically happen. When you've got my my wonderful girlfriend of first
[01:37:04] got my my wonderful girlfriend of first ads, assistant directors, you've got
[01:37:07] ads, assistant directors, you've got gaffers, you've got lighters, and also
[01:37:09] gaffers, you've got lighters, and also simulating light is insanely difficult.
[01:37:11] simulating light is insanely difficult. There are so many magical things that
[01:37:13] There are so many magical things that happen in creating visual images that
[01:37:15] happen in creating visual images that yeah, you could create a one minute long
[01:37:16] yeah, you could create a one minute long thing that might fool someone. How do
[01:37:18] thing that might fool someone. How do you practically turn that into a movie?
[01:37:20] you practically turn that into a movie? Because that movie, I forget what the
[01:37:22] Because that movie, I forget what the name is. There was a movie that claimed
[01:37:23] name is. There was a movie that claimed it aired at Can. It didn't. No one. It
[01:37:26] it aired at Can. It didn't. No one. It aired in the city of Can during the Can
[01:37:28] aired in the city of Can during the Can Film Festival. It was not at the film
[01:37:30] Film Festival. It was not at the film festival. When it comes to the practical
[01:37:31] festival. When it comes to the practical creation of actual things at the end of
[01:37:33] creation of actual things at the end of it versus magic tricks, the actual
[01:37:35] it versus magic tricks, the actual practical outcomes are not there. The
[01:37:37] practical outcomes are not there. The reason I keep coming back to the
[01:37:38] reason I keep coming back to the capabilities thing for the example is
[01:37:40] capabilities thing for the example is yeah, they can do better at tests, do
[01:37:41] yeah, they can do better at tests, do better number go up. When it comes to
[01:37:44] better number go up. When it comes to can this actually do distinct tasks you
[01:37:46] can this actually do distinct tasks you can rely on it, you can rely on it for
[01:37:48] can rely on it, you can rely on it for summaries. You can rely on it for
[01:37:50] summaries. You can rely on it for generations. The things it was doing,
[01:37:51] generations. The things it was doing, it's getting linearlyish better at. But
[01:37:54] it's getting linearlyish better at. But again, there's a ceiling to that. Like,
[01:37:57] again, there's a ceiling to that. Like, okay, so it gets really good at
[01:37:58] okay, so it gets really good at research. What does that actually mean?
[01:38:00] research. What does that actually mean? you've already kind of got the
[01:38:01] you've already kind of got the automation there. What is the next step
[01:38:03] automation there. What is the next step of that? Because training it to be more
[01:38:04] of that? Because training it to be more autonomous for example, that's not
[01:38:06] autonomous for example, that's not something that comes from training data.
[01:38:07] something that comes from training data. That is actually a new Gary Marcus a
[01:38:10] That is actually a new Gary Marcus a neuros symbolic. You actually need to
[01:38:12] neuros symbolic. You actually need to build a structure around the AI to make
[01:38:14] build a structure around the AI to make it work. And even then, it doesn't fix
[01:38:16] it work. And even then, it doesn't fix the
[01:38:16] the >> So you're saying that there will become
[01:38:18] >> So you're saying that there will become a point where the rate of improvement
[01:38:20] a point where the rate of improvement will plateau.
[01:38:21] will plateau. >> We're already there and stop.
[01:38:22] >> We're already there and stop. >> We've already hit that diminishing. Gary
[01:38:24] >> We've already hit that diminishing. Gary Marcus said this in 2022 as well. Do you
[01:38:26] Marcus said this in 2022 as well. Do you know there's lots of people listening
[01:38:27] know there's lots of people listening now that like they've had their
[01:38:29] now that like they've had their workflows completely transformed by
[01:38:31] workflows completely transformed by these tools? Have they?
[01:38:32] these tools? Have they? >> There'll be people. Yeah, there are.
[01:38:34] >> There'll be people. Yeah, there are. Yeah. The thing is, first of all, every
[01:38:36] Yeah. The thing is, first of all, every single one of them, did you pay for the
[01:38:37] single one of them, did you pay for the tokens? That's the thing. Did you pay
[01:38:39] tokens? That's the thing. Did you pay for the tokens? And also, how many
[01:38:41] for the tokens? And also, how many tokens did you burn? But putting all
[01:38:42] tokens did you burn? But putting all that aside, what workflows? Because if
[01:38:44] that aside, what workflows? Because if it's, yeah, I did a bunch of web
[01:38:45] it's, yeah, I did a bunch of web scraping or web searches. I'm just not
[01:38:47] scraping or web searches. I'm just not impressed. Did you make an entire
[01:38:49] impressed. Did you make an entire movie? No, you didn't. Is it
[01:38:51] movie? No, you didn't. Is it speeding up your coding? Yeah, I believe
[01:38:53] speeding up your coding? Yeah, I believe that. I've heard that from multiple
[01:38:54] that. I've heard that from multiple people. But again, how much can you
[01:38:57] people. But again, how much can you trust this?
[01:38:57] trust this? >> I think I'm I was getting at is, you
[01:39:00] >> I think I'm I was getting at is, you know, when in the moment of any
[01:39:02] know, when in the moment of any technological innovation, people they
[01:39:05] technological innovation, people they extrapolate linearly or they view it as
[01:39:09] extrapolate linearly or they view it as a static state, i.e. they think today is
[01:39:10] a static state, i.e. they think today is going to look like tomorrow or they
[01:39:12] going to look like tomorrow or they think it's going to get better in this
[01:39:13] think it's going to get better in this sort of straight line. But what we end
[01:39:14] sort of straight line. But what we end up seeing a lot of the time is this
[01:39:16] up seeing a lot of the time is this exponential improvement. All of the
[01:39:17] exponential improvement. All of the innovations we're talking about with you
[01:39:19] innovations we're talking about with you with like with compute and all that with
[01:39:21] with like with compute and all that with fast processes, those are hardware
[01:39:22] fast processes, those are hardware breakthroughs. The hardware breakthrough
[01:39:25] breakthroughs. The hardware breakthrough companies don't seem to be fixing the
[01:39:26] companies don't seem to be fixing the LLM problems despite the all the king's
[01:39:29] LLM problems despite the all the king's horses, all the king's men with what
[01:39:30] horses, all the king's men with what nine 10 generations of TPUs from Google
[01:39:33] nine 10 generations of TPUs from Google now. Broadcoms building stuff with open
[01:39:35] now. Broadcoms building stuff with open AI, their halapeno chip. And yet none of
[01:39:38] AI, their halapeno chip. And yet none of these people can just say, "Yeah, we're
[01:39:39] these people can just say, "Yeah, we're on the path to making this profitable."
[01:39:41] on the path to making this profitable." Because they can't. If we fix the
[01:39:42] Because they can't. If we fix the environmental problems and the
[01:39:43] environmental problems and the profitability situation, maybe I'd be
[01:39:46] profitability situation, maybe I'd be more generous with this stuff. But they
[01:39:48] more generous with this stuff. But they don't seem to be able to. And you talk
[01:39:51] don't seem to be able to. And you talk about these improvements and
[01:39:52] about these improvements and capabilities. There's a certain point at
[01:39:55] capabilities. There's a certain point at which I'm saying, "Okay, can it do even
[01:39:57] which I'm saying, "Okay, can it do even a tenth of the stuff they're promising?"
[01:39:59] a tenth of the stuff they're promising?" Sam the other week was saying it
[01:40:01] Sam the other week was saying it was going to be in like 6 months will be
[01:40:03] was going to be in like 6 months will be like a genie that you can ask wishes for
[01:40:05] like a genie that you can ask wishes for from like never watched
[01:40:07] from like never watched Aladdin. What's he talking about? Like
[01:40:09] Aladdin. What's he talking about? Like also the the genie was charming. Anyway,
[01:40:11] also the the genie was charming. Anyway, long story short, the promises do not
[01:40:14] long story short, the promises do not line up with the capabilities or the
[01:40:16] line up with the capabilities or the capability improvements. An exponential
[01:40:18] capability improvements. An exponential improvement
[01:40:19] improvement in software and software performance is
[01:40:23] in software and software performance is always a result of direct hardware
[01:40:25] always a result of direct hardware improvement. We have all the gifted
[01:40:27] improvement. We have all the gifted mathematicians, all the gifted software
[01:40:29] mathematicians, all the gifted software engineers, all the gifted hardware
[01:40:30] engineers, all the gifted hardware engineers. And where are we? Trillion
[01:40:33] engineers. And where are we? Trillion plus dollars in with the future great
[01:40:36] plus dollars in with the future great financial crisis and the world's
[01:40:37] financial crisis and the world's greatest marketing scop.
[01:40:39] greatest marketing scop. >> I just think in the future I do think
[01:40:40] >> I just think in the future I do think that all of the devices and the
[01:40:42] that all of the devices and the computers we use and the physical items
[01:40:44] computers we use and the physical items in our world will be more intelligent. I
[01:40:46] in our world will be more intelligent. I mean sure but is that LLMs
[01:40:48] mean sure but is that LLMs >> and that will be powered by the
[01:40:49] >> and that will be powered by the underlying AI infrastructure. It will be
[01:40:51] underlying AI infrastructure. It will be the more data data centers. It will be
[01:40:54] the more data data centers. It will be energy coming down.
[01:40:55] energy coming down. >> How does a GPU full data center
[01:40:59] >> How does a GPU full data center translate to a Nikon camera that can I
[01:41:04] translate to a Nikon camera that can I don't know even what you'd think think
[01:41:05] don't know even what you'd think think like because what is the thing we're
[01:41:07] like because what is the thing we're talking about here? Because the idea
[01:41:09] talking about here? Because the idea that devices will get smarter. Sure, I
[01:41:11] that devices will get smarter. Sure, I can see that. It's a very broad
[01:41:13] can see that. It's a very broad statement. I could see it happening.
[01:41:14] statement. I could see it happening. It's really kind of happening. What does
[01:41:16] It's really kind of happening. What does that have to do with the data centers?
[01:41:17] that have to do with the data centers? Cuz these data centers again are not
[01:41:19] Cuz these data centers again are not being built to make your consumer
[01:41:21] being built to make your consumer electronics smarter. They're not being
[01:41:23] electronics smarter. They're not being built for anything other than
[01:41:24] built for anything other than speculating on the ability to capture
[01:41:26] speculating on the ability to capture demand for generative AI services.
[01:41:28] demand for generative AI services. >> But it's not just generative AI. We went
[01:41:29] >> But it's not just generative AI. We went through that earlier.
[01:41:30] through that earlier. >> Yes. No, but those data centers, they
[01:41:31] >> Yes. No, but those data centers, they are being built for generative AI. They
[01:41:33] are being built for generative AI. They are not being built for anything else.
[01:41:35] are not being built for anything else. Would you consider generative AI to be
[01:41:37] Would you consider generative AI to be the fact that on Meta's earnings call
[01:41:39] the fact that on Meta's earnings call like a couple of weeks ago, Mark
[01:41:40] like a couple of weeks ago, Mark Zuckerberg said, "The big breakthrough
[01:41:42] Zuckerberg said, "The big breakthrough we've had, which has resulted in 15
[01:41:43] we've had, which has resulted in 15 basis points of increased retention, I
[01:41:46] basis points of increased retention, I believe he was referring to Instagram,
[01:41:48] believe he was referring to Instagram, is that we now take anything you post on
[01:41:51] is that we now take anything you post on social media and we run it through an AI
[01:41:53] social media and we run it through an AI to get full context of what it is." And
[01:41:55] to get full context of what it is." And because we can see guy sat in front of
[01:41:58] because we can see guy sat in front of me called Ed with blue shirt and coffee,
[01:42:01] me called Ed with blue shirt and coffee, we now can train the AI to serve whoever
[01:42:04] we now can train the AI to serve whoever wants blue shirt, Ed, and with coffee to
[01:42:06] wants blue shirt, Ed, and with coffee to the right user, which means people are
[01:42:08] the right user, which means people are retained longer because
[01:42:09] retained longer because >> it'sn't 15 basis points, like 0.15%.
[01:42:12] >> it'sn't 15 basis points, like 0.15%. >> Yeah, it's cool. But it makes a
[01:42:13] >> Yeah, it's cool. But it makes a difference at scale. It makes a big
[01:42:15] difference at scale. It makes a big difference at scale.
[01:42:16] difference at scale. >> Yeah. But 10 and something billion
[01:42:17] >> Yeah. But 10 and something billion dollars in and the best you've got is
[01:42:18] dollars in and the best you've got is 0.15%. If if he could be fight I mean
[01:42:22] 0.15%. If if he could be fight I mean how much of a difference because
[01:42:24] how much of a difference because >> there's a reason he's saying basis
[01:42:26] >> there's a reason he's saying basis points versus dollars
[01:42:28] points versus dollars >> because think about it like this if Mark
[01:42:30] >> because think about it like this if Mark Zuckerberg was
[01:42:32] Zuckerberg was >> I take your point about scale. No, I'm
[01:42:34] >> I take your point about scale. No, I'm saying the point I was making was that
[01:42:35] saying the point I was making was that that is another application of these
[01:42:38] that is another application of these data centers because it needs a data
[01:42:40] data centers because it needs a data center that is driving revenues, but
[01:42:43] center that is driving revenues, but also that's not out that's outside of us
[01:42:46] also that's not out that's outside of us thinking about just generating
[01:42:47] thinking about just generating >> and that's generative
[01:42:49] >> and that's generative model. Muse was it? Oh, Muse Spark is
[01:42:52] model. Muse was it? Oh, Muse Spark is their LLM. Gem is their generative ad
[01:42:54] their LLM. Gem is their generative ad model. Well, Muse then then that's them
[01:42:57] model. Well, Muse then then that's them doing the weird thing where it's like on
[01:42:58] doing the weird thing where it's like on Instagram and it's like Dave the cat.
[01:43:00] Instagram and it's like Dave the cat. Why is Dave the cat suffering? Like it's
[01:43:02] Why is Dave the cat suffering? Like it's the weird popup things. Meta is
[01:43:04] the weird popup things. Meta is god damn that company sucks. Like every
[01:43:06] god damn that company sucks. Like every time I think about how they've ruined
[01:43:07] time I think about how they've ruined that product. But that's the thing
[01:43:09] that product. But that's the thing though, again, why can't he just say
[01:43:10] though, again, why can't he just say with his whole chest, we've made a
[01:43:11] with his whole chest, we've made a couple billion. Why can't he say that?
[01:43:13] couple billion. Why can't he say that? Because he isn't. Because there's not
[01:43:15] Because he isn't. Because there's not actually a way of going, I spent all
[01:43:17] actually a way of going, I spent all this money. I spent 14 billion goddamn
[01:43:19] this money. I spent 14 billion goddamn dollars on scale Alexander Wong and I
[01:43:22] dollars on scale Alexander Wong and I made this much. They can't. It gets back
[01:43:25] made this much. They can't. It gets back to a very simple point of, hey, if it
[01:43:28] to a very simple point of, hey, if it was going well, you'd tell me how well
[01:43:29] was going well, you'd tell me how well it was going rather than, I don't know,
[01:43:32] it was going rather than, I don't know, doing this weird rain dance thing where
[01:43:34] doing this weird rain dance thing where you're like, well, if we move all the
[01:43:35] you're like, well, if we move all the pieces around in 3 years, theoretically,
[01:43:37] pieces around in 3 years, theoretically, this will happen.
[01:43:40] this will happen. I've done almost 700 interviews with
[01:43:42] I've done almost 700 interviews with some of the most interesting people in
[01:43:43] some of the most interesting people in the world. And one of the things you
[01:43:45] the world. And one of the things you learn, which is unexpected, is that
[01:43:46] learn, which is unexpected, is that vulnerability is the doorway to
[01:43:49] vulnerability is the doorway to connection. And after sitting here for 2
[01:43:51] connection. And after sitting here for 2 three hours with a guest, I feel a deep
[01:43:53] three hours with a guest, I feel a deep sense of connection to them. And as they
[01:43:55] sense of connection to them. And as they leave, what I get them to do is to write
[01:43:58] leave, what I get them to do is to write a question in the diary of a CEO. We've
[01:44:01] a question in the diary of a CEO. We've taken all of the questions from the
[01:44:03] taken all of the questions from the diary of a CEO. We have put the question
[01:44:07] diary of a CEO. We have put the question here on this card with the name of the
[01:44:09] here on this card with the name of the person that wrote it. So you can sit at
[01:44:11] person that wrote it. So you can sit at home as I do with my fiance and my
[01:44:13] home as I do with my fiance and my colleagues at work and other people in
[01:44:15] colleagues at work and other people in my life. Whenever we get a minute, we
[01:44:17] my life. Whenever we get a minute, we play the diio conversation cards and it
[01:44:20] play the diio conversation cards and it is incredible what happens. These are
[01:44:23] is incredible what happens. These are great if you're in a romantic
[01:44:24] great if you're in a romantic relationship and you want to connect
[01:44:25] relationship and you want to connect your partner more. These are also great
[01:44:27] your partner more. These are also great if you're in a team and you want to bond
[01:44:29] if you're in a team and you want to bond your team together. And I have to say
[01:44:30] your team together. And I have to say they're also great for families that
[01:44:32] they're also great for families that want to learn more about each other and
[01:44:34] want to learn more about each other and that need a good excuse to spend some
[01:44:36] that need a good excuse to spend some time in a digital world in the analog
[01:44:39] time in a digital world in the analog environment connecting human to human.
[01:44:41] environment connecting human to human. It is remarkable what the right question
[01:44:44] It is remarkable what the right question at the right time can do. Go to the
[01:44:46] at the right time can do. Go to the diary.com
[01:44:48] diary.com and you can get these conversation cards
[01:44:50] and you can get these conversation cards right now. There should be a button just
[01:44:53] right now. There should be a button just down below here. And if it says
[01:44:55] down below here. And if it says subscribed, you're already subscribed.
[01:44:56] subscribed, you're already subscribed. If it says subscriber, that means you're
[01:44:59] If it says subscriber, that means you're not yet. And if you're not subscribed,
[01:45:00] not yet. And if you're not subscribed, please could you do us a favor and hit
[01:45:01] please could you do us a favor and hit that button? It helps the show more than
[01:45:03] that button? It helps the show more than you know. And according to the
[01:45:04] you know. And according to the algorithm, you're someone that watches
[01:45:06] algorithm, you're someone that watches our show, but you haven't yet hit that
[01:45:07] our show, but you haven't yet hit that button. Thank you so much. I do think
[01:45:10] button. Thank you so much. I do think you're accurate and right when you talk
[01:45:12] you're accurate and right when you talk about the fact that there's a lot of
[01:45:13] about the fact that there's a lot of like is the word for gazy?
[01:45:14] like is the word for gazy? >> Yeah.
[01:45:15] >> Yeah. >> Where like there's a lot of people that
[01:45:16] >> Where like there's a lot of people that have spent a lot of money and they kind
[01:45:17] have spent a lot of money and they kind of shouldn't have spent it and they
[01:45:18] of shouldn't have spent it and they up and now they're thinking
[01:45:20] up and now they're thinking like we've spent all this invested money
[01:45:22] like we've spent all this invested money kind of like the metaverse was a bit of
[01:45:23] kind of like the metaverse was a bit of a
[01:45:24] a >> oh my god that was a bit of a joke.
[01:45:25] >> oh my god that was a bit of a joke. >> That's so weird.
[01:45:25] >> That's so weird. >> A lot of money spent. We kind of thought
[01:45:27] >> A lot of money spent. We kind of thought this dream was coming of this well I
[01:45:29] this dream was coming of this well I shouldn't say dream cuz it's not a dream
[01:45:30] shouldn't say dream cuz it's not a dream I've had but
[01:45:31] I've had but >> dream that they had.
[01:45:32] >> dream that they had. >> Yeah. This sort of virtual world and
[01:45:34] >> Yeah. This sort of virtual world and actually it never transpired and there's
[01:45:35] actually it never transpired and there's no sign that it will in the near term.
[01:45:37] no sign that it will in the near term. AI and the dotcom boom in this regard
[01:45:41] AI and the dotcom boom in this regard are the same. NFTTS were the same,
[01:45:43] are the same. NFTTS were the same, >> you know. So crypto, one could argue
[01:45:45] >> you know. So crypto, one could argue that a lot of the crypto industry was
[01:45:46] that a lot of the crypto industry was the same. It's weighing that is inflated
[01:45:48] the same. It's weighing that is inflated by the media. The difference is the
[01:45:50] by the media. The difference is the reason the metaverse and NFTs didn't
[01:45:52] reason the metaverse and NFTs didn't escape this was there weren't stocks to
[01:45:54] escape this was there weren't stocks to speculate on. There weren't big
[01:45:56] speculate on. There weren't big companies that you could invest in. They
[01:45:58] companies that you could invest in. They had re record earnings in 2021. There's
[01:46:00] had re record earnings in 2021. There's a bunch of money floating in the system
[01:46:02] a bunch of money floating in the system thanks to postcoid uh the PDC that
[01:46:04] thanks to postcoid uh the PDC that basically government federal money
[01:46:06] basically government federal money flowed in to the banks. There was a
[01:46:08] flowed in to the banks. There was a bunch of easy money zero interest free
[01:46:10] bunch of easy money zero interest free era money was easy to find. Then after
[01:46:11] era money was easy to find. Then after that there was the hangover. Growth
[01:46:13] that there was the hangover. Growth started to slow down dramatically. This
[01:46:15] started to slow down dramatically. This is actually my rockcom bubble theory
[01:46:16] is actually my rockcom bubble theory which is they don't have any hyperrowth
[01:46:18] which is they don't have any hyperrowth ideas anymore. So suddenly they started
[01:46:21] ideas anymore. So suddenly they started buying GPUs. And when they bought GPUs
[01:46:24] buying GPUs. And when they bought GPUs people went they're doing AI. Oh we
[01:46:26] people went they're doing AI. Oh we better buy the stock. And the stocks
[01:46:27] better buy the stock. And the stocks went on an incredible run. may like
[01:46:29] went on an incredible run. may like several hundred percent grow in the last
[01:46:31] several hundred percent grow in the last few years. the stock has grown by
[01:46:32] few years. the stock has grown by hundreds of percent. Despite zero proof
[01:46:35] hundreds of percent. Despite zero proof and because the media was just saying,
[01:46:38] and because the media was just saying, "Yeah, Meta's revenues growing because
[01:46:40] "Yeah, Meta's revenues growing because of AI, right? Microsoft's revenue is
[01:46:42] of AI, right? Microsoft's revenue is grown because of AI, right? The fugazi
[01:46:44] grown because of AI, right? The fugazi you're talking about was the fact that
[01:46:45] you're talking about was the fact that everyone just gave them credit in
[01:46:47] everyone just gave them credit in advance and now we're kind of getting to
[01:46:49] advance and now we're kind of getting to the point where it's like, hey, you
[01:46:50] the point where it's like, hey, you didn't spend that trillion dollars for
[01:46:52] didn't spend that trillion dollars for no reason, did you? Satcha Amy Amy Hood
[01:46:55] no reason, did you? Satcha Amy Amy Hood just going to take him out back, send
[01:46:56] just going to take him out back, send him to the glue factory or something?"
[01:46:58] him to the glue factory or something?" Like,
[01:46:58] Like, >> I do think there's overspending. I I
[01:47:00] >> I do think there's overspending. I I want to concede that but I doic
[01:47:02] want to concede that but I doic >> yeah no I do think there is and I think
[01:47:03] >> yeah no I do think there is and I think the reason why there's overspending Ed
[01:47:06] the reason why there's overspending Ed is I think there is something here
[01:47:09] is I think there is something here >> and what
[01:47:10] >> and what >> in terms of like I think there is pra p
[01:47:12] p p p p p p p p p p p p p p p p p p p practical uses for this technology and I
[01:47:14] practical uses for this technology and I think when people realize that through
[01:47:16] think when people realize that through history they go crazy because they want
[01:47:19] history they go crazy because they want to be the person that owns the
[01:47:20] to be the person that owns the opportunity.
[01:47:20] opportunity. >> I'm going to be honest I just I
[01:47:22] >> I'm going to be honest I just I fundamentally don't agree.
[01:47:23] fundamentally don't agree. >> You don't agree with which part you
[01:47:24] >> You don't agree with which part you >> I don't agree that this that the
[01:47:26] >> I don't agree that this that the speculation is a result of actual
[01:47:27] speculation is a result of actual demand. I don't believe it's suspect. I
[01:47:29] demand. I don't believe it's suspect. I don't think private credit is sinking
[01:47:31] don't think private credit is sinking hundreds of billions of dollars into AI
[01:47:32] hundreds of billions of dollars into AI because of actual demand. They are doing
[01:47:34] because of actual demand. They are doing it because they saw the biggest
[01:47:35] it because they saw the biggest companies in the world building data
[01:47:36] companies in the world building data centers making a ton of money from two
[01:47:38] centers making a ton of money from two companies they feed money and went I
[01:47:40] companies they feed money and went I want some of that money.
[01:47:41] want some of that money. >> I am saying that I do think there is
[01:47:43] >> I am saying that I do think there is value in the underlying technology. I
[01:47:45] value in the underlying technology. I think that and so I think I'm not saying
[01:47:47] think that and so I think I'm not saying how much value
[01:47:48] how much value >> right okay I actually I get your meaning
[01:47:50] >> right okay I actually I get your meaning that's fair.
[01:47:51] that's fair. >> I'm not saying it's proportionate to the
[01:47:52] >> I'm not saying it's proportionate to the investment. All I'm saying is that do
[01:47:54] investment. All I'm saying is that do you know what it's like? It's like if I
[01:47:56] you know what it's like? It's like if I take your example, the rot economy essay
[01:47:58] take your example, the rot economy essay that you wrote.
[01:47:58] that you wrote. >> Yeah.
[01:47:59] >> Yeah. >> Say that you're on a desert island and
[01:48:00] >> Say that you're on a desert island and then someone says they found a banana
[01:48:02] then someone says they found a banana tree,
[01:48:03] tree, >> right?
[01:48:04] >> right? >> And there's there's 10,000 people on the
[01:48:06] >> And there's there's 10,000 people on the island.
[01:48:07] island. >> Okay.
[01:48:08] >> Okay. >> They are going to stam peed
[01:48:10] >> They are going to stam peed towards where they think the banana tree
[01:48:12] towards where they think the banana tree is. They are going to claw each
[01:48:14] is. They are going to claw each other to pieces. And if if your essay
[01:48:16] other to pieces. And if if your essay here is right that there was desperation
[01:48:18] here is right that there was desperation cuz they hadn't found an innovation in a
[01:48:20] cuz they hadn't found an innovation in a while,
[01:48:20] while, >> maybe that explains it. Maybe there is a
[01:48:22] >> maybe that explains it. Maybe there is a bit of value here,
[01:48:23] bit of value here, >> right?
[01:48:23] >> right? >> And they're stam peeding and
[01:48:25] >> And they're stam peeding and killing each other and making irrational
[01:48:27] killing each other and making irrational decisions like hungry people would.
[01:48:29] decisions like hungry people would. >> I actually think we're then we actually
[01:48:31] >> I actually think we're then we actually agree. That is actually my point, which
[01:48:32] agree. That is actually my point, which is these three companies in Meta, their
[01:48:35] is these three companies in Meta, their main business lines are running out of
[01:48:36] main business lines are running out of growth. There's only so much they can
[01:48:38] growth. There's only so much they can grow. And indeed, in the next three and
[01:48:39] grow. And indeed, in the next three and a half years, analysts think that these
[01:48:40] a half years, analysts think that these two bastards, these two, OpenAI and
[01:48:42] two bastards, these two, OpenAI and Anthropic are going to spend over $400
[01:48:44] Anthropic are going to spend over $400 billion on these people alone,
[01:48:46] billion on these people alone, Microsoft, Google, and Amazon. And the
[01:48:48] Microsoft, Google, and Amazon. And the crazy thing is is that's a large part of
[01:48:50] crazy thing is is that's a large part of their future growth. And if this money
[01:48:51] their future growth. And if this money isn't spent, their growth slows down.
[01:48:54] isn't spent, their growth slows down. Okay,
[01:48:54] Okay, >> so your point about a bananas, I
[01:48:55] >> so your point about a bananas, I actually agree. That is the rockcom
[01:48:57] actually agree. That is the rockcom bubble, it's they don't have a new thing
[01:48:58] bubble, it's they don't have a new thing and they're desperate. And indeed, they
[01:49:00] and they're desperate. And indeed, they got rewarded for buying the GPUs. They
[01:49:03] got rewarded for buying the GPUs. They got when they bought these goddamn GPUs
[01:49:05] got when they bought these goddamn GPUs from Nvidia, all the markets went
[01:49:07] from Nvidia, all the markets went rockard overnight. They loved it. There
[01:49:09] rockard overnight. They loved it. There were stories about how they were sending
[01:49:11] were stories about how they were sending armored cars with the GPUs to Microsoft
[01:49:13] armored cars with the GPUs to Microsoft to make sure Microsoft got the GPUs. And
[01:49:15] to make sure Microsoft got the GPUs. And so everyone saw all that money flowing
[01:49:17] so everyone saw all that money flowing in. Even though they never disclosed AI
[01:49:19] in. Even though they never disclosed AI revenues, they saw the expenditures and
[01:49:20] revenues, they saw the expenditures and they went, "Well, I want to do what
[01:49:22] they went, "Well, I want to do what these people are doing. I want to get a
[01:49:24] these people are doing. I want to get a little of that money, don't I?"
[01:49:25] little of that money, don't I?" >> I think the area where we have a slight
[01:49:27] >> I think the area where we have a slight disagreement is that I think the
[01:49:30] disagreement is that I think the underlying technology has a lot more
[01:49:32] underlying technology has a lot more promise over the long term than you do.
[01:49:35] promise over the long term than you do. So the thing I want to push back on
[01:49:36] So the thing I want to push back on there is
[01:49:38] there is to have progress with AI just on a
[01:49:41] to have progress with AI just on a taking it in a vacuum to have progress
[01:49:43] taking it in a vacuum to have progress for these two companies to keep going
[01:49:45] for these two companies to keep going and to keep progressing they need to
[01:49:47] and to keep progressing they need to spend tens of billions of dollars a year
[01:49:49] spend tens of billions of dollars a year on training.
[01:49:51] on training. >> The only way that that can happen is if
[01:49:53] >> The only way that that can happen is if these companies and venture capitalists
[01:49:55] these companies and venture capitalists and private credit firms and Nvidia
[01:49:57] and private credit firms and Nvidia >> keep circulating money to them. So the
[01:49:59] >> keep circulating money to them. So the progress
[01:50:00] progress >> that we've got so far is entirely a
[01:50:02] >> that we've got so far is entirely a result of this circular system. So it
[01:50:04] result of this circular system. So it means that
[01:50:05] means that >> circular you talked about VCs there
[01:50:07] >> circular you talked about VCs there >> venture capitalists who are by the way
[01:50:09] >> venture capitalists who are by the way the majority of the funding that open
[01:50:12] the majority of the funding that open AAI got in the last 6 months came from
[01:50:14] AAI got in the last 6 months came from SoftBank Nvidia and Amazon
[01:50:16] SoftBank Nvidia and Amazon >> okay yeah
[01:50:17] >> okay yeah >> so just the point is is you're talking
[01:50:19] >> so just the point is is you're talking about progress continuing progress in
[01:50:21] about progress continuing progress in LLM can only continue as long as the
[01:50:24] LLM can only continue as long as the money keeps flowing once the money keep
[01:50:26] money keeps flowing once the money keep once the money stops flowing the
[01:50:28] once the money stops flowing the progress stops which
[01:50:29] progress stops which >> but isn't that most like early like
[01:50:30] >> but isn't that most like early like Spotify didn't make money for 20 years
[01:50:32] Spotify didn't make money for 20 years >> Spotify didn't lose 20.9 9 billion in
[01:50:34] >> Spotify didn't lose 20.9 9 billion in one year. They didn't need to raise $217
[01:50:37] one year. They didn't need to raise $217 billion in the space of 6 months.
[01:50:39] billion in the space of 6 months. >> Yeah. And Uber is another example.
[01:50:40] >> Yeah. And Uber is another example. >> $33 billion since inception before it
[01:50:42] >> $33 billion since inception before it became a messy kind of profitable.
[01:50:44] became a messy kind of profitable. Amazon Web Services between 2003 and
[01:50:46] Amazon Web Services between 2003 and 2015 when it became profitable. $29.7
[01:50:48] 2015 when it became profitable. $29.7 billion the scale. Yeah. That's the
[01:50:51] billion the scale. Yeah. That's the total capital expenditures and that's
[01:50:52] total capital expenditures and that's not just Amazon Web Services. That's the
[01:50:54] not just Amazon Web Services. That's the entire logistics operation normalized
[01:50:56] entire logistics operation normalized for inflation.
[01:50:57] for inflation. >> So they all lost money for a long period
[01:50:58] >> So they all lost money for a long period of time is the TLDDR.
[01:51:00] of time is the TLDDR. >> Yes. But the amount of money they lost
[01:51:02] >> Yes. But the amount of money they lost is
[01:51:04] is completely
[01:51:06] completely just magnitudes different on a level
[01:51:08] just magnitudes different on a level where these three
[01:51:09] where these three >> Can I argue then that the that's because
[01:51:12] >> Can I argue then that the that's because the potential of intelligence permeates
[01:51:15] the potential of intelligence permeates everything whereas Amazon at the time
[01:51:16] everything whereas Amazon at the time was like selling books
[01:51:18] was like selling books >> no
[01:51:18] >> no >> that was that was bringing retail online
[01:51:20] >> that was that was bringing retail online >> when Amazon web services grew it was
[01:51:22] >> when Amazon web services grew it was >> oh so cloud with Amazon web services the
[01:51:25] >> oh so cloud with Amazon web services the reason I bring that up going to repeat
[01:51:27] reason I bring that up going to repeat something but it's really important 2003
[01:51:29] something but it's really important 2003 it was founded
[01:51:30] it was founded >> and it was founded mostly because Amazon
[01:51:32] >> and it was founded mostly because Amazon as a growing online store needed
[01:51:34] as a growing online store needed hardcore infrastructure. 2006, I think,
[01:51:36] hardcore infrastructure. 2006, I think, is when they turned it client-f facing.
[01:51:38] is when they turned it client-f facing. I may be wrong on the dates there, but
[01:51:39] I may be wrong on the dates there, but 2015 was the year it became profitable.
[01:51:41] 2015 was the year it became profitable. >> Yeah.
[01:51:42] >> Yeah. >> The total capital expenditures
[01:51:43] >> The total capital expenditures normalized for inflation with $29.7
[01:51:46] normalized for inflation with $29.7 billion across that 12-year period.
[01:51:48] billion across that 12-year period. >> Yeah.
[01:51:49] >> Yeah. >> And yeah, it lost money, but
[01:51:51] >> And yeah, it lost money, but >> if we speak cold economics here, Amazon
[01:51:55] >> if we speak cold economics here, Amazon didn't have to go into the they were
[01:51:57] didn't have to go into the they were unprofitable in in a way, but their
[01:51:58] unprofitable in in a way, but their margins actually started improving
[01:52:00] margins actually started improving because AWS was a very margin heavy
[01:52:01] because AWS was a very margin heavy business. It was great.
[01:52:02] business. It was great. >> Yeah,
[01:52:03] >> Yeah, >> these these two Google cash flow
[01:52:06] >> these these two Google cash flow negative, Amazon cash flow negative.
[01:52:08] negative, Amazon cash flow negative. These businesses, the reason you liked
[01:52:10] These businesses, the reason you liked software businesses was they are meant
[01:52:12] software businesses was they are meant to be cash heavy asset light. These
[01:52:17] to be cash heavy asset light. These companies along with Meta have added
[01:52:19] companies along with Meta have added more than $700 billion of new property,
[01:52:22] more than $700 billion of new property, plants and equipment. So assets, data
[01:52:24] plants and equipment. So assets, data centers, GPUs in the last four years.
[01:52:26] centers, GPUs in the last four years. They have gone from being these cash
[01:52:28] They have gone from being these cash machines to these cash furnaces.
[01:52:32] machines to these cash furnaces. >> You said a second ago, this can only
[01:52:33] >> You said a second ago, this can only continue if if investors continue to
[01:52:36] continue if if investors continue to invest.
[01:52:36] invest. >> Yes.
[01:52:36] >> Yes. >> And I was saying I I think that
[01:52:38] >> And I was saying I I think that investors are used to pumping money into
[01:52:41] investors are used to pumping money into things that are burning cash. Your
[01:52:42] things that are burning cash. Your rebuttal to me sounds like well this is
[01:52:44] rebuttal to me sounds like well this is burning more cash than ever. And then so
[01:52:46] burning more cash than ever. And then so I would say well is the opportunity
[01:52:49] I would say well is the opportunity bigger than those other case studies you
[01:52:52] bigger than those other case studies you referenced like AWS? And one would say
[01:52:54] referenced like AWS? And one would say that the opportunity of intelligence
[01:52:59] that the opportunity of intelligence permeates everything. So the TAM the
[01:53:01] permeates everything. So the TAM the total addressable market is enormous.
[01:53:04] total addressable market is enormous. Maybe the revival back to me is about
[01:53:05] Maybe the revival back to me is about open source and all these kind of
[01:53:06] open source and all these kind of >> No, no, no. I I actually know what
[01:53:08] >> No, no, no. I I actually know what you're getting at. So what you were
[01:53:09] you're getting at. So what you were describing there is the argument that
[01:53:11] describing there is the argument that Sachinadella or Sam would make that the
[01:53:13] Sachinadella or Sam would make that the theoretical opportunity of large
[01:53:15] theoretical opportunity of large language models and I could have bought
[01:53:17] language models and I could have bought that into any 24 from them when
[01:53:19] that into any 24 from them when they were like, "Oh, we see the
[01:53:20] they were like, "Oh, we see the opportunity. We've gone way past the
[01:53:22] opportunity. We've gone way past the point at which you can rationally argue
[01:53:25] point at which you can rationally argue that LLMs need this much money. And when
[01:53:27] that LLMs need this much money. And when I say the money needs to keep flowing, I
[01:53:29] I say the money needs to keep flowing, I am talking these two compan Open AI just
[01:53:32] am talking these two compan Open AI just open AI Clammy Sam has said Wall Street
[01:53:36] open AI Clammy Sam has said Wall Street Journal and Isaagi reported a few weeks
[01:53:38] Journal and Isaagi reported a few weeks ago they plan to spend $750 billion on
[01:53:42] ago they plan to spend $750 billion on compute through 2030. I think they're
[01:53:44] compute through 2030. I think they're going to be dead before then, but $750
[01:53:47] going to be dead before then, but $750 billion.
[01:53:48] billion. That is an insane amount of money. That
[01:53:50] That is an insane amount of money. That is crazy
[01:53:51] is crazy >> and [laughter]
[01:53:52] >> and [laughter] a large chunk of that is training. So
[01:53:54] a large chunk of that is training. So when I say progress, I mean literally to
[01:53:56] when I say progress, I mean literally to make the models better at stuff requires
[01:53:58] make the models better at stuff requires billions of dollars invested just in
[01:53:59] billions of dollars invested just in data
[01:54:01] data and also tens of billions of dollars of
[01:54:02] and also tens of billions of dollars of taking that data. And so training
[01:54:04] taking that data. And so training training is actually a really
[01:54:05] training is actually a really interesting thing because when you think
[01:54:08] interesting thing because when you think of like for Jake and Troy my trainers
[01:54:10] of like for Jake and Troy my trainers when I train with them when I lift with
[01:54:12] when I train with them when I lift with them I have a defined thing and when I
[01:54:13] them I have a defined thing and when I do it and I eat right muscles get bigger
[01:54:15] do it and I eat right muscles get bigger they would. And here's the thing. When
[01:54:17] they would. And here's the thing. When you train with an LLM, you're
[01:54:19] you train with an LLM, you're experimenting each and this is not
[01:54:21] experimenting each and this is not actually a hit on the companies because
[01:54:22] actually a hit on the companies because they're still trying to work out how to
[01:54:24] they're still trying to work out how to do the thing because putting aside how I
[01:54:27] do the thing because putting aside how I feel like they're trying to innovate. I
[01:54:28] feel like they're trying to innovate. I think there are people at these
[01:54:29] think there are people at these companies that actually want to do
[01:54:30] companies that actually want to do something interesting. It's costing too
[01:54:32] something interesting. It's costing too much money. So once the money tap turns
[01:54:34] much money. So once the money tap turns off, the money won't be there to buy the
[01:54:37] off, the money won't be there to buy the data or feed the data into the GPUs. Put
[01:54:39] data or feed the data into the GPUs. Put aside all the thoughts I have, just the
[01:54:41] aside all the thoughts I have, just the raw capital to get them this far has
[01:54:43] raw capital to get them this far has cost increasingly larger amounts of
[01:54:45] cost increasingly larger amounts of money and increasingly larger amounts of
[01:54:47] money and increasingly larger amounts of training money for training runs that
[01:54:49] training money for training runs that sometimes can fail. GPT5 was meant to be
[01:54:53] sometimes can fail. GPT5 was meant to be this panacea for the AI industry. They
[01:54:55] this panacea for the AI industry. They had at least one training run that cost
[01:54:57] had at least one training run that cost half a billion dollars and did nothing.
[01:54:59] half a billion dollars and did nothing. And that's the thing. If we are thinking
[01:55:01] And that's the thing. If we are thinking about progress in a in a vacuum, they
[01:55:03] about progress in a in a vacuum, they need so much more money just to maybe
[01:55:06] need so much more money just to maybe get somewhere. There's no guarantee.
[01:55:07] get somewhere. There's no guarantee. There's never any guarantee, but there's
[01:55:08] There's never any guarantee, but there's a reason that Google and Amazon are cash
[01:55:10] a reason that Google and Amazon are cash flow negative now. There's a reason why
[01:55:12] flow negative now. There's a reason why Oracle's probably going to die as a
[01:55:14] Oracle's probably going to die as a result of OpenAI because Oracle's future
[01:55:16] result of OpenAI because Oracle's future depends on OpenAI spending $300 billion
[01:55:18] depends on OpenAI spending $300 billion over 5 years.
[01:55:19] over 5 years. >> It's absolutely fascinating because I
[01:55:21] >> It's absolutely fascinating because I was just reading through a list of
[01:55:22] was just reading through a list of quotes from the big CEOs of AI companies
[01:55:25] quotes from the big CEOs of AI companies to see what they would rebuttle you.
[01:55:27] to see what they would rebuttle you. >> Yeah.
[01:55:28] >> Yeah. >> And they're all basically saying the
[01:55:29] >> And they're all basically saying the same thing. They're all saying, this is
[01:55:31] same thing. They're all saying, this is actual an exact quote from Sundar who is
[01:55:34] actual an exact quote from Sundar who is the CEO of Google. He says the risk of
[01:55:36] the CEO of Google. He says the risk of underinvesting is dramatically greater
[01:55:39] underinvesting is dramatically greater than the risk of overinvesting.
[01:55:43] than the risk of overinvesting. And you go down, you go through this,
[01:55:44] And you go down, you go through this, you know, Andy Jasse, CEO of Amazon,
[01:55:46] you know, Andy Jasse, CEO of Amazon, we're not investing approximately 200
[01:55:48] we're not investing approximately 200 billion in capex in 2026 on a hunch.
[01:55:51] billion in capex in 2026 on a hunch. We're not going to be conservative in
[01:55:53] We're not going to be conservative in how we play this. We're investing to be
[01:55:56] how we play this. We're investing to be the meaningful leader and our future
[01:55:58] the meaningful leader and our future business operating income and free cash
[01:56:00] business operating income and free cash flow will be much larger because of this
[01:56:02] flow will be much larger because of this investment. Then Mark Zuckerberg, CE of
[01:56:05] investment. Then Mark Zuckerberg, CE of Meta, says we'll continue to invest
[01:56:07] Meta, says we'll continue to invest aggressively in infrastructure to meet
[01:56:08] aggressively in infrastructure to meet the demand. I'd rather risk building
[01:56:11] the demand. I'd rather risk building capacity before it's needed than being
[01:56:13] capacity before it's needed than being late. Makes me think of Shrek with L
[01:56:15] late. Makes me think of Shrek with L Farquad. Some of you may die, but that's
[01:56:17] Farquad. Some of you may die, but that's a risk I'm willing to accept. It's like,
[01:56:20] a risk I'm willing to accept. It's like, you know, I'm just going to spend all
[01:56:21] you know, I'm just going to spend all this money. You can't fire me cuz Mark
[01:56:22] this money. You can't fire me cuz Mark Zuckerberg can't be fired due to the
[01:56:24] Zuckerberg can't be fired due to the unique board situation he's got going.
[01:56:26] unique board situation he's got going. So yeah, he's just going to piss the
[01:56:27] So yeah, he's just going to piss the money away and hope he's right. And I
[01:56:28] money away and hope he's right. And I know from the people who know it matter,
[01:56:30] know from the people who know it matter, he's not right. The thing is, why might
[01:56:32] he's not right. The thing is, why might you be wrong?
[01:56:33] you be wrong? >> I mean, this is the thing. The AI people
[01:56:35] >> I mean, this is the thing. The AI people who claim this is going to be the
[01:56:36] who claim this is going to be the biggest, strongest thing in the world,
[01:56:37] biggest, strongest thing in the world, did they ever get that? I I mean this
[01:56:39] did they ever get that? I I mean this like
[01:56:40] like >> it's a good question because it's like
[01:56:41] >> it's a good question because it's like they don't. And the thing is, what would
[01:56:43] they don't. And the thing is, what would it take for me to be wrong? A bunch of
[01:56:44] it take for me to be wrong? A bunch of hardware breakthroughs to make this
[01:56:45] hardware breakthroughs to make this profitable. A bunch of
[01:56:47] profitable. A bunch of >> question new mathemat because the thing
[01:56:48] >> question new mathemat because the thing is
[01:56:49] is >> when it comes to being a critic or a
[01:56:51] >> when it comes to being a critic or a skeptic,
[01:56:52] skeptic, >> you are put on the hot seat. Not the
[01:56:54] >> you are put on the hot seat. Not the people spending a trillion dollars, not
[01:56:55] people spending a trillion dollars, not the people promising the world. The
[01:56:57] the people promising the world. The person the the with a blog is
[01:56:59] person the the with a blog is the one who's like me. Trust me. If they
[01:57:01] the one who's like me. Trust me. If they came here, they'd be on the hot seat,
[01:57:02] came here, they'd be on the hot seat, too. Trust me.
[01:57:03] too. Trust me. >> Oh, I Oh, they they won't talk to me.
[01:57:05] >> Oh, I Oh, they they won't talk to me. Don't know why, Steve. They don't know.
[01:57:07] Don't know why, Steve. They don't know. It's cuz I call him Clammy Sammy. Um
[01:57:09] It's cuz I call him Clammy Sammy. Um >> I think it's cuz my guests are quite
[01:57:11] >> I think it's cuz my guests are quite quite critical that I don't think Solman
[01:57:13] quite critical that I don't think Solman wants to come here.
[01:57:13] wants to come here. >> Mr. Orman, go on Steve show. Do it. But
[01:57:16] >> Mr. Orman, go on Steve show. Do it. But this is the thing like of course they're
[01:57:17] this is the thing like of course they're going to say that. And also, if they
[01:57:19] going to say that. And also, if they thought they were right, I don't think
[01:57:21] thought they were right, I don't think they do anymore. If I was in their shoes
[01:57:23] they do anymore. If I was in their shoes and I thought that this was an
[01:57:24] and I thought that this was an existential thing, sure. But it gets
[01:57:25] existential thing, sure. But it gets back to the rocom bubble which is yeah
[01:57:27] back to the rocom bubble which is yeah this is the last thing they've got.
[01:57:29] this is the last thing they've got. >> But I really want to know that question.
[01:57:30] >> But I really want to know that question. It was one of the questions I was really
[01:57:31] It was one of the questions I was really excited to ask you which is you have a
[01:57:33] excited to ask you which is you have a different opinion. We said this at the
[01:57:34] different opinion. We said this at the top. You have a very different opinion
[01:57:36] top. You have a very different opinion from a lot of people. I would categorize
[01:57:38] from a lot of people. I would categorize the the two most popular opinions as
[01:57:41] the the two most popular opinions as >> uh AI is going to hurt everybody and
[01:57:43] >> uh AI is going to hurt everybody and it's going to be catastrophic and we
[01:57:44] it's going to be catastrophic and we need to stop.
[01:57:45] need to stop. >> Yeah.
[01:57:45] >> Yeah. >> The other opinion is age of abundance is
[01:57:47] >> The other opinion is age of abundance is going to be amazing. Let us crack on.
[01:57:49] going to be amazing. Let us crack on. yours is different from both of those
[01:57:51] yours is different from both of those which is as you said in your words it's
[01:57:53] which is as you said in your words it's a con and it's and there's no real
[01:57:56] a con and it's and there's no real underlying value in the technology and
[01:57:57] underlying value in the technology and it's overhyped.
[01:57:58] it's overhyped. >> Yes.
[01:57:59] >> Yes. >> And there's way too much spending. I
[01:58:00] >> And there's way too much spending. I mean a few people agree on the spending
[01:58:02] mean a few people agree on the spending part but the other part. So with you
[01:58:04] part but the other part. So with you it's one of probably the first person
[01:58:05] it's one of probably the first person that I've spoken to that's had this
[01:58:07] that I've spoken to that's had this opinion.
[01:58:08] opinion. >> So how what would it take for you to
[01:58:11] >> So how what would it take for you to change your mind about what you believe
[01:58:14] change your mind about what you believe here? There would need to be a hardware
[01:58:16] here? There would need to be a hardware breakthrough that reduced the cost by
[01:58:18] breakthrough that reduced the cost by like a thousand but it would have to be
[01:58:20] like a thousand but it would have to be just a dramatic breakthrough that is not
[01:58:22] just a dramatic breakthrough that is not happening just to be clear because
[01:58:24] happening just to be clear because they've all been trying. So it's the
[01:58:25] they've all been trying. So it's the cost for you that would have to change.
[01:58:27] cost for you that would have to change. >> It's the cost and it's also the data
[01:58:28] >> It's the cost and it's also the data centers. I think the way they're
[01:58:30] centers. I think the way they're building the data centers is reckless
[01:58:31] building the data centers is reckless and damaging to communities. The fact
[01:58:33] and damaging to communities. The fact that you have communities like in
[01:58:34] that you have communities like in violent New Jersey where the residents
[01:58:36] violent New Jersey where the residents like I don't want this but the planning
[01:58:37] like I don't want this but the planning boards vote for it because they're all I
[01:58:39] boards vote for it because they're all I assume having chummy lunches with the
[01:58:41] assume having chummy lunches with the people doing it. I think the use of gas
[01:58:43] people doing it. I think the use of gas turbines is disgraceful. I the
[01:58:46] turbines is disgraceful. I the water situation I'm not super well read
[01:58:48] water situation I'm not super well read on, so I'm not going to wait into it,
[01:58:49] on, so I'm not going to wait into it, but the use of gas turbines and behind
[01:58:50] but the use of gas turbines and behind the meter power is reckless and damaging
[01:58:53] the meter power is reckless and damaging to communities. The noise that these
[01:58:54] to communities. The noise that these things make and also generative AI is
[01:58:57] things make and also generative AI is this egregious pornographic
[01:59:01] this egregious pornographic demonstration of how unfair the world
[01:59:03] demonstration of how unfair the world is. Regular people try and get a loan
[01:59:05] is. Regular people try and get a loan for a business, a random business. They
[01:59:06] for a business, a random business. They want I have a good idea. They go to a
[01:59:08] want I have a good idea. They go to a bank, a bank of town, go
[01:59:10] bank, a bank of town, go themselves. They'll say, "I'm not g you
[01:59:11] themselves. They'll say, "I'm not g you going to make a store that sells stuff.
[01:59:13] going to make a store that sells stuff. Screw you. You want to build a data
[01:59:15] Screw you. You want to build a data center? You Jensen Hang will back you.
[01:59:18] center? You Jensen Hang will back you. Jensen Hong will give you 25% residual
[01:59:20] Jensen Hong will give you 25% residual value. You want to build a regular
[01:59:22] value. You want to build a regular business that's even profitable?
[01:59:24] business that's even profitable? you. No, a venture capitalist won't give
[01:59:25] you. No, a venture capitalist won't give you the money. Something that's just
[01:59:27] you the money. Something that's just growing steadily, but it's profitable.
[01:59:28] growing steadily, but it's profitable. Screw that. No, I need 10 100x return.
[01:59:31] Screw that. No, I need 10 100x return. Try and get a mortgage. You have to give
[01:59:34] Try and get a mortgage. You have to give the bank a full colonic. But you want to
[01:59:36] the bank a full colonic. But you want to get money for Jensen Hong to buy some
[01:59:38] get money for Jensen Hong to buy some GPUs? He'll give you a contract.
[01:59:39] GPUs? He'll give you a contract. Corewave is a great example. C Neocloud,
[01:59:42] Corewave is a great example. C Neocloud, which is just a company that builds data
[01:59:44] which is just a company that builds data centers and puts GPUs and rent them to
[01:59:45] centers and puts GPUs and rent them to people. Nvidia, one of their first
[01:59:48] people. Nvidia, one of their first investors in 2023, signed a $1.3 billion
[01:59:52] investors in 2023, signed a $1.3 billion contract to rent back their GPUs from
[01:59:55] contract to rent back their GPUs from Core. So that Core go to a bank and go,
[01:59:57] Core. So that Core go to a bank and go, I got a customer. Yeah, it's the guy I'm
[01:59:59] I got a customer. Yeah, it's the guy I'm buying the GPUs from with the debt I'm
[02:00:01] buying the GPUs from with the debt I'm getting from you. If you want to buy
[02:00:03] getting from you. If you want to buy GPUs, it's open season. If you want to
[02:00:05] GPUs, it's open season. If you want to live a regular life where you build a
[02:00:06] live a regular life where you build a regular business or buy a house, highest
[02:00:09] regular business or buy a house, highest interest rates ever. Screw you. Up
[02:00:11] interest rates ever. Screw you. Up yours. Yeah, you need to show us way
[02:00:13] yours. Yeah, you need to show us way more than that. I don't trust you
[02:00:15] more than that. I don't trust you regular folks. But if you're an
[02:00:16] regular folks. But if you're an unprofitable Neocloud, you get billions
[02:00:19] unprofitable Neocloud, you get billions from Jensen. It doesn't matter.
[02:00:21] from Jensen. It doesn't matter. >> It's so interesting. You It's
[02:00:23] >> It's so interesting. You It's interesting because you are the first
[02:00:24] interesting because you are the first person that I've spoken to that has that
[02:00:26] person that I've spoken to that has that opinion.
[02:00:26] opinion. >> I am prouser. Let's take another myth.
[02:00:30] >> I am prouser. Let's take another myth. AI will be conscious. Mhm. So
[02:00:34] AI will be conscious. Mhm. So super intelligence, artificial general
[02:00:36] super intelligence, artificial general intelligence, these are theories. Anyone
[02:00:40] intelligence, these are theories. Anyone saying this stuff will become this is
[02:00:42] saying this stuff will become this is just guessing and does not have proof.
[02:00:45] just guessing and does not have proof. >> Okay.
[02:00:45] >> Okay. >> And like that's really it.
[02:00:47] >> And like that's really it. >> Okay.
[02:00:47] >> Okay. >> Okay. Let's take another myth.
[02:00:50] >> Okay. Let's take another myth. AI systems are already blackmailing and
[02:00:53] AI systems are already blackmailing and escaping control. So this is a really
[02:00:55] escaping control. So this is a really specific one. Anthropic. There's
[02:00:57] specific one. Anthropic. There's actually two. Open AAI's GPT 3.5. I
[02:01:01] actually two. Open AAI's GPT 3.5. I realize this is more than the sentence.
[02:01:02] realize this is more than the sentence. I apologize.
[02:01:04] I apologize. In their system card, and a bunch of
[02:01:06] In their system card, and a bunch of media outlets covered this, saying that
[02:01:08] media outlets covered this, saying that OpenAI's model blackmailed a task rabbit
[02:01:10] OpenAI's model blackmailed a task rabbit into solving a capture. What actually
[02:01:13] into solving a capture. What actually happened was a user of GPT doing the
[02:01:18] happened was a user of GPT doing the experiment
[02:01:19] experiment got it to generate things to say to a
[02:01:22] got it to generate things to say to a task rabbit to make a task rabbit do
[02:01:24] task rabbit to make a task rabbit do stuff.
[02:01:24] stuff. >> A task rabbit
[02:01:25] >> A task rabbit >> as in a person that you rent, not even
[02:01:27] >> as in a person that you rent, not even to do a capture. It's something you rent
[02:01:29] to do a capture. It's something you rent to like nail a picture up in your
[02:01:31] to like nail a picture up in your apartment. It's an insane example. This
[02:01:33] apartment. It's an insane example. This was covered as if these things
[02:01:34] was covered as if these things blackmailed someone and and it and they
[02:01:36] blackmailed someone and and it and they specifically said, "Yeah, we prompted it
[02:01:38] specifically said, "Yeah, we prompted it to do this." And also the other note was
[02:01:40] to do this." And also the other note was that yeah, AI systems can't do
[02:01:42] that yeah, AI systems can't do autonomous stuff like this. Then there
[02:01:44] autonomous stuff like this. Then there was this other one where Anthropic said,
[02:01:45] was this other one where Anthropic said, "Oh yeah, a model was blackmailing
[02:01:48] "Oh yeah, a model was blackmailing someone saying that if you don't do
[02:01:50] someone saying that if you don't do this, I'll email proof that you slept
[02:01:52] this, I'll email proof that you slept with someone else other than your wife."
[02:01:53] with someone else other than your wife." I think it was what actually happened
[02:01:55] I think it was what actually happened was Anthropic explicitly trained a model
[02:01:57] was Anthropic explicitly trained a model to do this and then prompted it to
[02:01:59] to do this and then prompted it to blackmail.
[02:02:01] blackmail. This keeps happening and the media just
[02:02:04] This keeps happening and the media just slop slot me up. I don't need no
[02:02:06] slop slot me up. I don't need no thoughts. Put the story in the bag. And
[02:02:08] thoughts. Put the story in the bag. And it's frustrating because it scares
[02:02:10] it's frustrating because it scares people. Put aside the fact it's wrong.
[02:02:12] people. Put aside the fact it's wrong. It's scary. It's scary to people. people
[02:02:15] It's scary. It's scary to people. people living their lives who have to work
[02:02:17] living their lives who have to work longer hours to make less money and
[02:02:19] longer hours to make less money and their money doesn't go far and they turn
[02:02:20] their money doesn't go far and they turn on the news and there's some
[02:02:22] on the news and there's some being like, "Yeah, you should be
[02:02:23] being like, "Yeah, you should be terrified it blackmailed someone."
[02:02:26] terrified it blackmailed someone." >> But this is this is so counterintuitive
[02:02:28] >> But this is this is so counterintuitive of their interest to some degree and
[02:02:30] of their interest to some degree and they've experienced it backfire.
[02:02:32] they've experienced it backfire. >> Well, they have now like it's it's
[02:02:33] >> Well, they have now like it's it's literally backfired.
[02:02:34] literally backfired. >> It's backfired. Eric Schmidt getting
[02:02:36] >> It's backfired. Eric Schmidt getting booed at a commencement speech by 8,000
[02:02:38] booed at a commencement speech by 8,000 people every time he said the word AI.
[02:02:40] people every time he said the word AI. But I mean this is this is I mean these
[02:02:43] But I mean this is this is I mean these serious are being attacked at home.
[02:02:45] serious are being attacked at home. >> Yeah. Which sucks. Which is
[02:02:46] >> Yeah. Which sucks. Which is >> terrible. I must be clear like you
[02:02:48] >> terrible. I must be clear like you dislike the don't hurt people.
[02:02:50] dislike the don't hurt people. >> Yeah. Don't don't attack people at home.
[02:02:51] >> Yeah. Don't don't attack people at home. But but the point here is that that
[02:02:54] But but the point here is that that narrative is backfiring in a big big way
[02:02:56] narrative is backfiring in a big big way for them. I don't think they saw it
[02:02:58] for them. I don't think they saw it coming because you have to remember you
[02:03:00] coming because you have to remember you mentioned regulation earlier. These tech
[02:03:02] mentioned regulation earlier. These tech companies have been glazed for their
[02:03:04] companies have been glazed for their entire existence. Travis Kick's like oh
[02:03:06] entire existence. Travis Kick's like oh what? People don't like me now. And it's
[02:03:08] what? People don't like me now. And it's because Uber was a horribly run place
[02:03:10] because Uber was a horribly run place and he was kind of a monster. Also tons
[02:03:12] and he was kind of a monster. Also tons of articles about how great Uber was at
[02:03:14] of articles about how great Uber was at the time. The point I'm making is these
[02:03:15] the time. The point I'm making is these companies are not used to push back.
[02:03:17] companies are not used to push back. They thought what would happen I believe
[02:03:19] They thought what would happen I believe just guessing. They thought they do this
[02:03:20] just guessing. They thought they do this scary stuff and they would just get
[02:03:22] scary stuff and they would just get floods of money and everyone would just
[02:03:24] floods of money and everyone would just be like I kneel before you. I'll do
[02:03:25] be like I kneel before you. I'll do whatever you want. They didn't expect I
[02:03:28] whatever you want. They didn't expect I think what has I I agree this has
[02:03:31] think what has I I agree this has backfired on them because they were in
[02:03:33] backfired on them because they were in articulate. They're disconnected from
[02:03:35] articulate. They're disconnected from regular people. Samman drives a $5
[02:03:37] regular people. Samman drives a $5 million car around San Francisco. So
[02:03:39] million car around San Francisco. So that that man's doing it like 9 miles an
[02:03:41] that that man's doing it like 9 miles an hour. It's hilarious. But these people
[02:03:43] hour. It's hilarious. But these people are disconnected from everyone else. So
[02:03:45] are disconnected from everyone else. So they don't they don't experience real
[02:03:46] they don't they don't experience real problems, so they can't build the
[02:03:47] problems, so they can't build the solutions for them. And they think,
[02:03:49] solutions for them. And they think, well, if we scare people into doing what
[02:03:50] well, if we scare people into doing what we want, that'll work, right? It didn't.
[02:03:53] we want, that'll work, right? It didn't. They was all of this blackmail stuff was
[02:03:55] They was all of this blackmail stuff was an attempt to make it mystic. It was a
[02:03:57] an attempt to make it mystic. It was a mysticism attempt. It was to make it
[02:03:59] mysticism attempt. It was to make it seem like this unknowable, impossible to
[02:04:01] seem like this unknowable, impossible to control, just this powerful thing. But
[02:04:03] control, just this powerful thing. But we're the only ones. We are the o only
[02:04:06] we're the only ones. We are the o only us only these two angels could possibly
[02:04:09] us only these two angels could possibly control the beast we've created.
[02:04:11] control the beast we've created. >> This is this is quite a controversial
[02:04:12] >> This is this is quite a controversial statement but I think that for some
[02:04:15] statement but I think that for some reason I trust Dario a little bit more
[02:04:18] reason I trust Dario a little bit more because I think he's been the most
[02:04:19] because I think he's been the most balanced in his writing about the risk
[02:04:21] balanced in his writing about the risk profile.
[02:04:22] profile. >> I
[02:04:23] >> I >> whereas the others they they seem to
[02:04:26] >> whereas the others they they seem to kind of move with the wind.
[02:04:27] kind of move with the wind. >> I I do you know
[02:04:29] >> I I do you know >> I get what you mean. The reason I don't
[02:04:31] >> I get what you mean. The reason I don't like Dario is Daario was doing the scare
[02:04:33] like Dario is Daario was doing the scare tactics thing when he worked at OpenAI
[02:04:35] tactics thing when he worked at OpenAI when GPT2 came out say it's too scary to
[02:04:37] when GPT2 came out say it's too scary to release. He's also gone on television
[02:04:40] release. He's also gone on television and given AI psychosis to Axios being
[02:04:42] and given AI psychosis to Axios being like 50% of jobs are going to go away
[02:04:45] like 50% of jobs are going to go away because of AI.
[02:04:46] because of AI. >> What I respect is the consistency. He's
[02:04:49] >> What I respect is the consistency. He's now being attacked by them.
[02:04:51] now being attacked by them. >> Good.
[02:04:51] >> Good. >> Um but the thing is sorry I mean let me
[02:04:54] >> Um but the thing is sorry I mean let me clarify the word attack. Darian is being
[02:04:56] clarify the word attack. Darian is being verbally attacked by Silicon Valley and
[02:04:59] verbally attacked by Silicon Valley and you know if Silicon Valley if powerful
[02:05:02] you know if Silicon Valley if powerful people in Silicon Valley are attacking
[02:05:03] people in Silicon Valley are attacking someone.
[02:05:04] someone. >> Four months ago he wasn't though. They
[02:05:06] >> Four months ago he wasn't though. They were all saying he was the smartest boy
[02:05:07] were all saying he was the smartest boy ever.
[02:05:08] ever. >> The point I want to make there as well
[02:05:08] >> The point I want to make there as well is again wow you're so scared of how
[02:05:10] is again wow you're so scared of how powerful this is. You're so scared of
[02:05:12] powerful this is. You're so scared of it. It's so scary. What are you doing
[02:05:13] it. It's so scary. What are you doing about it? Oh nothing. Like it's just
[02:05:15] about it? Oh nothing. Like it's just like what are you doing? Well we have an
[02:05:17] like what are you doing? Well we have an alignment team. So does every AI lab.
[02:05:19] alignment team. So does every AI lab. Well I guess open AI cycles through
[02:05:21] Well I guess open AI cycles through those really quickly. Here's the thing.
[02:05:22] those really quickly. Here's the thing. If I'm Dario Amade, I'm sitting there
[02:05:24] If I'm Dario Amade, I'm sitting there going, I'm scared of all things changing
[02:05:26] going, I'm scared of all things changing and I thought I had made a thing that
[02:05:28] and I thought I had made a thing that would eliminate all jobs, I'd be
[02:05:30] would eliminate all jobs, I'd be terrified. I'd be walking around with
[02:05:32] terrified. I'd be walking around with like like a 10 ton weight on my back.
[02:05:34] like like a 10 ton weight on my back. The show, the responsibility, the fact
[02:05:37] The show, the responsibility, the fact he doesn't, the fact he wants to be this
[02:05:39] he doesn't, the fact he wants to be this weird elder statesman that's too scared
[02:05:40] weird elder statesman that's too scared to hold Sam Orman's hand at an event
[02:05:43] to hold Sam Orman's hand at an event just makes me believe that he's just
[02:05:44] just makes me believe that he's just saying it because it's convenient and
[02:05:46] saying it because it's convenient and he'll wind that back as he kind of
[02:05:47] he'll wind that back as he kind of already has whenever it's convenient for
[02:05:50] already has whenever it's convenient for him. I think Open AAI and Anthropic are
[02:05:52] him. I think Open AAI and Anthropic are basically the same level of Bad Company.
[02:05:53] basically the same level of Bad Company. I think Anthropic is more cultlike. I
[02:05:56] I think Anthropic is more cultlike. I think it's so weird like Jack Clark over
[02:05:59] think it's so weird like Jack Clark over there, one of the co-founders. That fell
[02:06:00] there, one of the co-founders. That fell used to be at the register. He used to
[02:06:01] used to be at the register. He used to be one of the most critical journalists
[02:06:02] be one of the most critical journalists ever. Now he's it's like like something
[02:06:05] ever. Now he's it's like like something took over him because they talk of these
[02:06:07] took over him because they talk of these things in these high fluent terms. But
[02:06:08] things in these high fluent terms. But then again, maybe the people at
[02:06:09] then again, maybe the people at anthropic buy their Maybe some of
[02:06:11] anthropic buy their Maybe some of the people at OpenAI buy their I
[02:06:12] the people at OpenAI buy their I don't know. So going back to the central
[02:06:14] don't know. So going back to the central question we asked at the top here was
[02:06:15] question we asked at the top here was what would have to be the case for you
[02:06:17] what would have to be the case for you to look back and say do you know what I
[02:06:18] to look back and say do you know what I was wrong in 2026 and you said to me it
[02:06:20] was wrong in 2026 and you said to me it would be mainly that the cost of
[02:06:24] would be mainly that the cost of production around AI drops dramatically
[02:06:27] production around AI drops dramatically >> and it would have to also do insane
[02:06:29] >> and it would have to also do insane amounts of stuff it does it would have
[02:06:30] amounts of stuff it does it would have to be a truly autonomous
[02:06:32] to be a truly autonomous >> it would have to continue its
[02:06:33] >> it would have to continue its improvement in terms of capability.
[02:06:35] improvement in terms of capability. >> It would have to be a different product.
[02:06:36] >> It would have to be a different product. It would have to be it would have to be
[02:06:37] It would have to be it would have to be indistinguishable from magic. And the
[02:06:39] indistinguishable from magic. And the reason they have these high standards is
[02:06:40] reason they have these high standards is they set them.
[02:06:41] they set them. >> Okay. Fair. It's interesting as well
[02:06:43] >> Okay. Fair. It's interesting as well because all these myths and all these
[02:06:45] because all these myths and all these conversations, it's about technology,
[02:06:46] conversations, it's about technology, but it's also it's an information war.
[02:06:48] but it's also it's an information war. It's literally
[02:06:50] It's literally narrative versus narrative. Everyone
[02:06:53] narrative versus narrative. Everyone trying to escape the financials,
[02:06:55] trying to escape the financials, everyone trying to actually escape what
[02:06:56] everyone trying to actually escape what the models can do. And the big thing I
[02:06:58] the models can do. And the big thing I always say about AI boosters is if I
[02:07:00] always say about AI boosters is if I could regulate them, I'd regulate them.
[02:07:02] could regulate them, I'd regulate them. They can't speak in the future tense
[02:07:03] They can't speak in the future tense anymore. Just you got to talk about
[02:07:05] anymore. Just you got to talk about today, mate. You get two weeks in the
[02:07:07] today, mate. You get two weeks in the future, Max. Because if they were
[02:07:09] future, Max. Because if they were constrained to what was happening today,
[02:07:11] constrained to what was happening today, it they would sound like insane people.
[02:07:13] it they would sound like insane people. >> Yeah. No, I think yeah, most I guess
[02:07:14] >> Yeah. No, I think yeah, most I guess most technology companies would at the
[02:07:16] most technology companies would at the time. Like Uber would sound insane.
[02:07:18] time. Like Uber would sound insane. Amazon was
[02:07:19] Amazon was >> Uber was basically the difference.
[02:07:20] >> Uber was basically the difference. >> They were pissing money though, weren't
[02:07:21] >> They were pissing money though, weren't they?
[02:07:21] they? >> They were pissing money away, but the
[02:07:23] >> They were pissing money away, but the unit economics were the same just
[02:07:24] unit economics were the same just subsidized. So you were still getting a
[02:07:27] subsidized. So you were still getting a service from A to B and paying a much
[02:07:30] service from A to B and paying a much lower cost. It wasn't like you paid Uber
[02:07:33] lower cost. It wasn't like you paid Uber 200 sorry 20 bucks a month and you could
[02:07:35] 200 sorry 20 bucks a month and you could get 500 miles of Uber and then one day
[02:07:37] get 500 miles of Uber and then one day you started paying by the mile cuz
[02:07:38] you started paying by the mile cuz that's what's happening with this.
[02:07:39] that's what's happening with this. >> Have they they've changed their business
[02:07:41] >> Have they they've changed their business model for customers like me now so that
[02:07:44] model for customers like me now so that I have to buy credits.
[02:07:45] I have to buy credits. >> No. So you well kind of with
[02:07:47] >> No. So you well kind of with >> they asked me the other day. So with the
[02:07:49] >> they asked me the other day. So with the anthropics fable model with some
[02:07:52] anthropics fable model with some accounts you have to pay for usage and
[02:07:54] accounts you have to pay for usage and also adoption of fable has been pretty
[02:07:56] also adoption of fable has been pretty low because of this because of the cost
[02:07:57] low because of this because of the cost but with enterprises so companies over
[02:08:00] but with enterprises so companies over 150 people you have to pay by the token
[02:08:02] 150 people you have to pay by the token now or per million token.
[02:08:04] now or per million token. >> Oh so they are moving to a token.
[02:08:05] >> Oh so they are moving to a token. >> Yeah. But when they did that everyone
[02:08:06] >> Yeah. But when they did that everyone went from being like this is the most
[02:08:08] went from being like this is the most impressive thing ever to being like
[02:08:11] impressive thing ever to being like >> it's always we got to control these
[02:08:12] >> it's always we got to control these costs. Uber's COO said as Andrew
[02:08:15] costs. Uber's COO said as Andrew McDonald I think he said that it's
[02:08:16] McDonald I think he said that it's getting hard to justify cuz it's hard to
[02:08:19] getting hard to justify cuz it's hard to connect spending money on tokens to
[02:08:21] connect spending money on tokens to actual useful outcomes.
[02:08:22] actual useful outcomes. >> He said the thing like he said the
[02:08:24] >> He said the thing like he said the actual thing I've been saying and it's
[02:08:26] actual thing I've been saying and it's so we're in an AI bubble.
[02:08:27] so we're in an AI bubble. >> Yes.
[02:08:28] >> Yes. >> And when will when this AI bubble
[02:08:30] >> And when will when this AI bubble collapses so much of the economy is
[02:08:32] collapses so much of the economy is resting upon it.
[02:08:34] resting upon it. >> Yeah.
[02:08:34] >> Yeah. >> It's going to have downstream
[02:08:35] >> It's going to have downstream consequences. So I got two questions for
[02:08:37] consequences. So I got two questions for you. I guess the first question is are
[02:08:38] you. I guess the first question is are we in an AI bubble and what happens when
[02:08:40] we in an AI bubble and what happens when the bubble pops?
[02:08:41] the bubble pops? >> Yes. And it's it depends. So the big
[02:08:46] >> Yes. And it's it depends. So the big thing that people say is, "Oh, we'll get
[02:08:47] thing that people say is, "Oh, we'll get bailed out. Donald Trump scared of
[02:08:49] bailed out. Donald Trump scared of Donald Trump." Here's the problem with
[02:08:50] Donald Trump." Here's the problem with this.
[02:08:52] this. It isn't just an AI bubble. It's the
[02:08:54] It isn't just an AI bubble. It's the rockcom bubble. So the AI bubble
[02:08:56] rockcom bubble. So the AI bubble collapsing will probably be this company
[02:08:59] collapsing will probably be this company running out of money. Open AI.
[02:09:01] running out of money. Open AI. >> And the thing is with Open AI is they
[02:09:03] >> And the thing is with Open AI is they were meant to go public this year and
[02:09:05] were meant to go public this year and now it's been pushed to next year a week
[02:09:07] now it's been pushed to next year a week and a half after I released their
[02:09:08] and a half after I released their auditive financials. Wonder where that
[02:09:09] auditive financials. Wonder where that was. Um, but they've delayed to next
[02:09:11] was. Um, but they've delayed to next year. Sarah Frier, the CFO, has now
[02:09:13] year. Sarah Frier, the CFO, has now said, "Well, they'll do it earlier than
[02:09:15] said, "Well, they'll do it earlier than 2027 or 2027." Great answer there.
[02:09:18] 2027 or 2027." Great answer there. >> For anyone that doesn't understand what
[02:09:20] >> For anyone that doesn't understand what going public means, that means joining
[02:09:21] going public means, that means joining the stock market. And at such a time
[02:09:23] the stock market. And at such a time when you join the stock market, your
[02:09:24] when you join the stock market, your investors can finally sell their equity
[02:09:27] investors can finally sell their equity that they got for investing in the
[02:09:29] that they got for investing in the company when it was private. So often
[02:09:32] company when it was private. So often times companies will flirt with the idea
[02:09:34] times companies will flirt with the idea of we'll go public someday soon because
[02:09:37] of we'll go public someday soon because investors will have a moment in their
[02:09:39] investors will have a moment in their head where they'll get their money back
[02:09:40] head where they'll get their money back at a return. So you kind of need to if
[02:09:43] at a return. So you kind of need to if you're in these guys shoes, you kind of
[02:09:45] you're in these guys shoes, you kind of need to be flirting with going public or
[02:09:46] need to be flirting with going public or investors won't want to invest.
[02:09:48] investors won't want to invest. >> Open AAI up until this point has been a
[02:09:50] >> Open AAI up until this point has been a private company and their last funding
[02:09:51] private company and their last funding round they were valued at $865 billion.
[02:09:55] round they were valued at $865 billion. Now when they tried to go public, New
[02:09:57] Now when they tried to go public, New York Times Mike Isaac reported this.
[02:10:00] York Times Mike Isaac reported this. They tried to list well they wanted to
[02:10:02] They tried to list well they wanted to go at a set a 1 trillion valuation.
[02:10:05] go at a set a 1 trillion valuation. Apparently their advisor said no don't
[02:10:08] Apparently their advisor said no don't do that. That is very bad for a number
[02:10:11] do that. That is very bad for a number of reasons. One open AI needs perpetual
[02:10:13] of reasons. One open AI needs perpetual amounts of money. They raised $122
[02:10:14] amounts of money. They raised $122 billion this year. Most of it's crossed.
[02:10:16] billion this year. Most of it's crossed. There's some left but they are going to
[02:10:19] There's some left but they are going to need to raise at least hundred billion a
[02:10:20] need to raise at least hundred billion a year just to survive. If they can't go
[02:10:23] year just to survive. If they can't go public they will have to raise another
[02:10:24] public they will have to raise another funding round. The problem is it's going
[02:10:26] funding round. The problem is it's going to be difficult to raise at even the
[02:10:28] to be difficult to raise at even the same one they raise that. They're
[02:10:29] same one they raise that. They're probably going to have to take a flat.
[02:10:31] probably going to have to take a flat. So the same amount. Exactly. But they
[02:10:34] So the same amount. Exactly. But they need money. They need money so bad.
[02:10:36] need money. They need money so bad. Amazon sent them $35 billion that was
[02:10:38] Amazon sent them $35 billion that was meant to be contingent on them going
[02:10:40] meant to be contingent on them going public early.
[02:10:42] public early. >> They did that because they need the
[02:10:43] >> They did that because they need the money. Now, OpenAI is the kind of
[02:10:46] money. Now, OpenAI is the kind of catastrophe center here because
[02:10:48] catastrophe center here because Anthropic is likely going to beat it to
[02:10:49] Anthropic is likely going to beat it to go public. And once Anthropic goes
[02:10:51] go public. And once Anthropic goes public, it'll be borderline impossible
[02:10:53] public, it'll be borderline impossible for Open AI to do so because Anthropic,
[02:10:55] for Open AI to do so because Anthropic, an unprofitable, unsustainable AI lab,
[02:10:57] an unprofitable, unsustainable AI lab, but a better business that's growing
[02:10:59] but a better business that's growing faster than Open AI's. I believe they
[02:11:01] faster than Open AI's. I believe they have a ceiling. They're eventually going
[02:11:02] have a ceiling. They're eventually going to face predition, too. I think sometime
[02:11:04] to face predition, too. I think sometime in 2027, things are going to start
[02:11:06] in 2027, things are going to start running out of steam. Because the thing
[02:11:07] running out of steam. Because the thing I said earlier, the only way these
[02:11:09] I said earlier, the only way these models get better is if you feed more
[02:11:11] models get better is if you feed more money, tens of billions of dollars into
[02:11:12] money, tens of billions of dollars into them.
[02:11:12] them. >> So, you think OpenAI runs out of steam
[02:11:14] >> So, you think OpenAI runs out of steam in 2027?
[02:11:15] in 2027? >> I think they're already running out of
[02:11:16] >> I think they're already running out of steam. Yeah. But I think they run out of
[02:11:18] steam. Yeah. But I think they run out of cash. You think they run out of cash?
[02:11:20] cash. You think they run out of cash? Yes. And the sequence of events here
[02:11:21] Yes. And the sequence of events here will be they they go out and try and
[02:11:23] will be they they go out and try and raise
[02:11:24] raise >> and they have trouble raising another
[02:11:25] >> and they have trouble raising another round. I think maybe Invidia props them
[02:11:27] round. I think maybe Invidia props them up a little. Maybe Private Credit,
[02:11:29] up a little. Maybe Private Credit, Blackstone, Black Rockck and the like
[02:11:31] Blackstone, Black Rockck and the like the ones and the reason that Private
[02:11:32] the ones and the reason that Private Credit is getting involved. So asset
[02:11:34] Credit is getting involved. So asset managers is because they're investing in
[02:11:35] managers is because they're investing in the data centers and they know this
[02:11:37] the data centers and they know this company's most of the data center
[02:11:38] company's most of the data center demand.
[02:11:39] demand. >> Okay. So they run out of steam in 2027
[02:11:41] >> Okay. So they run out of steam in 2027 according to you.
[02:11:41] according to you. >> Yep. And maybe they try if they bum rush
[02:11:43] >> Yep. And maybe they try if they bum rush to go public they're going to have worse
[02:11:44] to go public they're going to have worse economics than anthropic. They're going
[02:11:46] economics than anthropic. They're going to get savage. it. We work was a great
[02:11:48] to get savage. it. We work was a great example. Another SoftBank classic. Now,
[02:11:51] example. Another SoftBank classic. Now, I think Open AI collapses, there are
[02:11:53] I think Open AI collapses, there are many different ways it could happen.
[02:11:54] many different ways it could happen. There are many different ways it could
[02:11:56] There are many different ways it could end. But the crucial thing is is that
[02:11:58] end. But the crucial thing is is that there are multiple companies that are
[02:12:00] there are multiple companies that are existentially tied to OpenAI. SoftBank,
[02:12:03] existentially tied to OpenAI. SoftBank, one of the largest companies in the
[02:12:04] one of the largest companies in the Japanese stock market, a holding company
[02:12:06] Japanese stock market, a holding company with lots of investments. They have on
[02:12:09] with lots of investments. They have on paper about hundred billion worth of
[02:12:11] paper about hundred billion worth of OpenAI stock. If they can't go public,
[02:12:13] OpenAI stock. If they can't go public, they can't do diddly squat with that.
[02:12:15] they can't do diddly squat with that. And so Soft Bank's future, their ability
[02:12:18] And so Soft Bank's future, their ability to continue paying the people around
[02:12:19] to continue paying the people around them and existing as a business relies
[02:12:22] them and existing as a business relies on their ability to continually
[02:12:23] on their ability to continually liquidate funds to be to take the things
[02:12:26] liquidate funds to be to take the things they've invested in and have value from
[02:12:28] they've invested in and have value from them either by selling the stock or
[02:12:29] them either by selling the stock or taking loans out on the stock. If OpenAI
[02:12:31] taking loans out on the stock. If OpenAI can't go public, SoftBank can't do that.
[02:12:34] can't go public, SoftBank can't do that. SoftBank probably won't run out of
[02:12:36] SoftBank probably won't run out of money, but we're going to see one of the
[02:12:37] money, but we're going to see one of the largest holding companies in the world
[02:12:38] largest holding companies in the world become much smaller. We will also see
[02:12:41] become much smaller. We will also see Amazon, Google, and Microsoft have to
[02:12:44] Amazon, Google, and Microsoft have to restate guidance. they will have to say
[02:12:46] restate guidance. they will have to say actually we don't think we're going to
[02:12:47] actually we don't think we're going to grow as fast
[02:12:48] grow as fast >> and what happens then
[02:12:50] >> and what happens then >> well I think we enter a tech depression
[02:12:52] >> well I think we enter a tech depression because the rockcom bubble the core of
[02:12:53] because the rockcom bubble the core of my theory is that they're out of
[02:12:56] my theory is that they're out of hyperrowth ideas but the market doesn't
[02:12:58] hyperrowth ideas but the market doesn't think so the reason they're so
[02:13:00] think so the reason they're so maniacally spending is because buying AI
[02:13:04] maniacally spending is because buying AI GPUs allows them to kick the can further
[02:13:06] GPUs allows them to kick the can further allows them to say we're still doing
[02:13:07] allows them to say we're still doing something we're working on AI don't
[02:13:09] something we're working on AI don't think too hard and also their current
[02:13:10] think too hard and also their current businesses are still growing their
[02:13:12] businesses are still growing their current businesses will eventually slow
[02:13:14] current businesses will eventually slow there's only so many price increases.
[02:13:16] there's only so many price increases. There's only so many tweaks to ads. Only
[02:13:18] There's only so many tweaks to ads. Only so many tweaks to Google search. Only so
[02:13:20] so many tweaks to Google search. Only so only so many ways that Amazon can screw
[02:13:22] only so many ways that Amazon can screw merchants. So in that tech depression,
[02:13:25] merchants. So in that tech depression, which you think it might be triggered in
[02:13:28] which you think it might be triggered in 2027, is that a cascading downstream
[02:13:31] 2027, is that a cascading downstream economic depression? Because the stock
[02:13:33] economic depression? Because the stock market is heavily dependent on these
[02:13:35] market is heavily dependent on these companies. The stock market sees a
[02:13:37] companies. The stock market sees a pullback, investors stop investing, they
[02:13:39] pullback, investors stop investing, they get panicked.
[02:13:40] get panicked. >> Yes. I think that because
[02:13:42] >> Yes. I think that because >> what's the sort of downstream
[02:13:43] >> what's the sort of downstream consequence the sort of domino effect
[02:13:44] consequence the sort of domino effect >> there's so much to imagine that it's
[02:13:46] >> there's so much to imagine that it's difficult to capture everything but
[02:13:48] difficult to capture everything but there are a few things that worry me
[02:13:50] there are a few things that worry me first of all a ton of American money
[02:13:52] first of all a ton of American money just regular people's money retail
[02:13:53] just regular people's money retail investors are in these companies and
[02:13:55] investors are in these companies and they bought into the magnificent 7
[02:13:56] they bought into the magnificent 7 thinking the number go up forever is the
[02:13:59] thinking the number go up forever is the largest company on the Fortune 500 and
[02:14:01] largest company on the Fortune 500 and NASDAQ as well and like 7 to 8% of the
[02:14:04] NASDAQ as well and like 7 to 8% of the S&P 500 that company when in when the
[02:14:08] S&P 500 that company when in when the bottom falls out from Nvidia and we
[02:14:09] bottom falls out from Nvidia and we haven't really got into it but Nvidia is
[02:14:10] haven't really got into it but Nvidia is doing the most circular of financing,
[02:14:12] doing the most circular of financing, feeding companies money so that they can
[02:14:14] feeding companies money so that they can raise debt to buy more GPUs. I think
[02:14:16] raise debt to buy more GPUs. I think Nvidia's revenue could go 50 to 70%
[02:14:19] Nvidia's revenue could go 50 to 70% down. I think that Nvidia could put
[02:14:20] down. I think that Nvidia could put Nvidia back in 2022 was making
[02:14:22] Nvidia back in 2022 was making singledigit billion dollars.
[02:14:23] singledigit billion dollars. >> And what happens though, I'm thinking
[02:14:25] >> And what happens though, I'm thinking about like Jenny and Dave that are
[02:14:26] about like Jenny and Dave that are watching this right now and they are
[02:14:28] watching this right now and they are just normal people
[02:14:29] just normal people >> with normal jobs.
[02:14:30] >> with normal jobs. >> People's retirements are going to
[02:14:32] >> People's retirements are going to contract severely and I don't believe
[02:14:34] contract severely and I don't believe they're going to return to those values.
[02:14:36] they're going to return to those values. And I think that because so much of the
[02:14:38] And I think that because so much of the value of the S&P 500 and Russell 1000
[02:14:41] value of the S&P 500 and Russell 1000 index comes from these four companies
[02:14:43] index comes from these four companies and the rest of the magnificent 7. So
[02:14:44] and the rest of the magnificent 7. So Apple, Tesla, Meta as well. And the
[02:14:48] Apple, Tesla, Meta as well. And the thing is I don't know what happens after
[02:14:50] thing is I don't know what happens after that because venture capital has also
[02:14:53] that because venture capital has also more than half of venture capital last
[02:14:54] more than half of venture capital last year went into AI. I think most venture
[02:14:56] year went into AI. I think most venture capital investments in AI are going to
[02:14:58] capital investments in AI are going to zero because when it comes to building a
[02:15:00] zero because when it comes to building a company on top of an LLM, all of those
[02:15:02] company on top of an LLM, all of those are unprofitable too. And the thing is
[02:15:05] are unprofitable too. And the thing is LLM companies have not really been
[02:15:06] LLM companies have not really been acquired. The exception being Cursible
[02:15:08] acquired. The exception being Cursible by Elon Musk for the coding side, but
[02:15:11] by Elon Musk for the coding side, but you have Cognition, which is just
[02:15:13] you have Cognition, which is just another LLM company raising a $26
[02:15:15] another LLM company raising a $26 billion valuation. That means that
[02:15:17] billion valuation. That means that company has to go public cuz who's
[02:15:19] company has to go public cuz who's buying a company at $26 billion other
[02:15:20] buying a company at $26 billion other than Elon Musk. And there were rumors
[02:15:22] than Elon Musk. And there were rumors that Elon Musk was trying to buy them as
[02:15:24] that Elon Musk was trying to buy them as well. Is Elon Musk just going to pick
[02:15:25] well. Is Elon Musk just going to pick off every like LLM company like going to
[02:15:28] off every like LLM company like going to TJ Maxx for AI? Like Jesus
[02:15:29] TJ Maxx for AI? Like Jesus Christ.
[02:15:30] Christ. >> So is that a recession you're
[02:15:31] >> So is that a recession you're describing? It is a recession, but it's
[02:15:33] describing? It is a recession, but it's also a depression within people's
[02:15:35] also a depression within people's retirements. Like I'm talking about 20,
[02:15:37] retirements. Like I'm talking about 20, 30, 40% off the top of these companies
[02:15:40] 30, 40% off the top of these companies stock value.
[02:15:40] stock value. >> Economic contractions, recessions
[02:15:42] >> Economic contractions, recessions consistently lead to job losses and
[02:15:43] consistently lead to job losses and rising unemployment. When an economy
[02:15:45] rising unemployment. When an economy contracts, the mechanism driving job
[02:15:46] contracts, the mechanism driving job losses typically follows a predictable
[02:15:48] losses typically follows a predictable sequence. Falling demand, consumers and
[02:15:51] sequence. Falling demand, consumers and businesses spend less money, causing
[02:15:52] businesses spend less money, causing revenues across most industries to drop.
[02:15:54] revenues across most industries to drop. margin compression. With lower revenue
[02:15:56] margin compression. With lower revenue and often fixed overhead costs like rent
[02:15:59] and often fixed overhead costs like rent or debt, corporate profit shrink, and
[02:16:00] or debt, corporate profit shrink, and lastly, cost cutting measures to survive
[02:16:02] lastly, cost cutting measures to survive or protect profit margins, businesses
[02:16:03] or protect profit margins, businesses freeze hiring, reduce hours, and resort
[02:16:05] freeze hiring, reduce hours, and resort to layoffs. Yes, that's that would all
[02:16:08] to layoffs. Yes, that's that would all happen. But the thing is, we're talking
[02:16:10] happen. But the thing is, we're talking about equity values dropping and we're
[02:16:12] about equity values dropping and we're talking about there not really being a
[02:16:14] talking about there not really being a home for that value or that money.
[02:16:16] home for that value or that money. [snorts] So much is riding on these
[02:16:18] [snorts] So much is riding on these companies, but you can't bail it out.
[02:16:21] companies, but you can't bail it out. You can theoretically bail out OpenAI. I
[02:16:23] You can theoretically bail out OpenAI. I don't think it happens. You could pump
[02:16:24] don't think it happens. You could pump these dogs full of money and keep them
[02:16:27] these dogs full of money and keep them alive for a bit, but at some point
[02:16:28] alive for a bit, but at some point they're going to have to start. They
[02:16:29] they're going to have to start. They have between these two companies,
[02:16:30] have between these two companies, Anthropic and Open AI, you have $1.1
[02:16:33] Anthropic and Open AI, you have $1.1 trillion of commitments.
[02:16:35] trillion of commitments. >> Just OpenAI.
[02:16:37] >> Just OpenAI. >> Oracle is building 7.1 gawatt of data
[02:16:39] >> Oracle is building 7.1 gawatt of data centers. So over $400 billion worth just
[02:16:42] centers. So over $400 billion worth just for OpenAI. There is not a customer on
[02:16:44] for OpenAI. There is not a customer on Earth. And Oracle's revenue has been
[02:16:46] Earth. And Oracle's revenue has been flat the last 15 years when you adjust
[02:16:48] flat the last 15 years when you adjust for inflation. Without Open AI, Oracle
[02:16:50] for inflation. Without Open AI, Oracle dies. So you think open AAI is going to
[02:16:52] dies. So you think open AAI is going to crash and run out of money and that's
[02:16:53] crash and run out of money and that's going to cause this domino effect across
[02:16:54] going to cause this domino effect across these other big tech companies which is
[02:16:56] these other big tech companies which is going to impact the stock market and
[02:16:58] going to impact the stock market and impact the broader economy.
[02:16:59] impact the broader economy. >> Yes. And also the tens of thousands of
[02:17:01] >> Yes. And also the tens of thousands of people that will be laid off from the
[02:17:02] people that will be laid off from the tech sector. But also the venture
[02:17:04] tech sector. But also the venture capital thing is significant because
[02:17:06] capital thing is significant because venture capital has been having one of
[02:17:08] venture capital has been having one of the most historic
[02:17:11] the most historic bad runs in history since 2018. The
[02:17:15] bad runs in history since 2018. The average return from venture capital
[02:17:17] average return from venture capital total value put in. So the amount of
[02:17:18] total value put in. So the amount of money you get back for your dollar is
[02:17:19] money you get back for your dollar is between8 and 1.21 meaning for every
[02:17:22] between8 and 1.21 meaning for every dollar you invest you get 80 cents to
[02:17:24] dollar you invest you get 80 cents to $120
[02:17:24] $120 >> paper gains.
[02:17:26] >> paper gains. >> Well no that's just actual g like actual
[02:17:28] >> Well no that's just actual g like actual returns. Paper gains they'll give you
[02:17:29] returns. Paper gains they'll give you but even then internal rate return which
[02:17:31] but even then internal rate return which is a whole separate thing even that's
[02:17:32] is a whole separate thing even that's not very happy. But long story short
[02:17:35] not very happy. But long story short very simple venture capital is not
[02:17:37] very simple venture capital is not making money come out. Venture capital
[02:17:39] making money come out. Venture capital is not actually providing returns.
[02:17:41] is not actually providing returns. >> They're celebrating paper gains.
[02:17:42] >> They're celebrating paper gains. >> They're celebrating paper gains
[02:17:43] >> They're celebrating paper gains >> and they're raising off paper gains.
[02:17:45] >> and they're raising off paper gains. >> Mhm. And actually paper gains I mean
[02:17:46] >> Mhm. And actually paper gains I mean just being able to say oh look the
[02:17:48] just being able to say oh look the valuation of anthropic went up. So
[02:17:50] valuation of anthropic went up. So that's
[02:17:50] that's >> but that's that's what Google and Amazon
[02:17:51] >> but that's that's what Google and Amazon were doing. Google's last quarter they
[02:17:53] were doing. Google's last quarter they boosted their net profits profits on
[02:17:56] boosted their net profits profits on paper by $99 billion because of the
[02:17:59] paper by $99 billion because of the increased value of their SpaceX holding
[02:18:00] increased value of their SpaceX holding and their anthropic holding. And again
[02:18:04] and their anthropic holding. And again the fact that this is happening is
[02:18:05] the fact that this is happening is insane and the fact it's not a scandal
[02:18:07] insane and the fact it's not a scandal is insane but we live in this culture I
[02:18:09] is insane but we live in this culture I guess. But everyone is really benefiting
[02:18:13] guess. But everyone is really benefiting right now. Oh, it's really that it's
[02:18:14] right now. Oh, it's really that it's that great tweet. It's like when you're
[02:18:16] that great tweet. It's like when you're reaping, it's like, "Yeah, yeah,
[02:18:17] reaping, it's like, "Yeah, yeah, this rocks." Sewing. Ah, This
[02:18:19] this rocks." Sewing. Ah, This sucks. Because right now, they're all
[02:18:21] sucks. Because right now, they're all like, "Yeah, all the speculative gains
[02:18:22] like, "Yeah, all the speculative gains are awesome. The paper gains are
[02:18:24] are awesome. The paper gains are awesome. The theoreticals of anthropic
[02:18:26] awesome. The theoreticals of anthropic being worth $2 trillion. Wow. The
[02:18:27] being worth $2 trillion. Wow. The articles we can write, the promises we
[02:18:29] articles we can write, the promises we can make. Then when the rubber meets the
[02:18:31] can make. Then when the rubber meets the road, it's going to be pretty rough on
[02:18:33] road, it's going to be pretty rough on them because the valuation of Amazon,
[02:18:37] them because the valuation of Amazon, Google, Microsoft, and Meta is based on
[02:18:38] Google, Microsoft, and Meta is based on this idea that they will grow eternally,
[02:18:40] this idea that they will grow eternally, that they will grow forever. If that
[02:18:42] that they will grow forever. If that changes, to quote Ed Elson from ProfitG
[02:18:44] changes, to quote Ed Elson from ProfitG Markets again, it's this. They're all
[02:18:45] Markets again, it's this. They're all doing Botox right now. They're sinking
[02:18:47] doing Botox right now. They're sinking money into it to make themselves feel
[02:18:48] money into it to make themselves feel young again and the market believes
[02:18:50] young again and the market believes them. When the market doesn't, we're not
[02:18:52] them. When the market doesn't, we're not just talking about a depression. I'm
[02:18:53] just talking about a depression. I'm talking about the market valuing them
[02:18:55] talking about the market valuing them like airlines and saying, "Yeah, you're
[02:18:57] like airlines and saying, "Yeah, you're real big and you make money off your
[02:18:59] real big and you make money off your existing products, but guess what? You
[02:19:01] existing products, but guess what? You don't have new You're just going
[02:19:03] don't have new You're just going to be doing this forever and we're going
[02:19:04] to be doing this forever and we're going to value you as such."
[02:19:05] to value you as such." >> So, if it's Jenny and Dave, should they
[02:19:08] >> So, if it's Jenny and Dave, should they do anything differently? Should they be
[02:19:10] do anything differently? Should they be conserving money? If there's a recession
[02:19:11] conserving money? If there's a recession or depression coming, should they be a
[02:19:13] or depression coming, should they be a little bit more conservative? Should
[02:19:14] little bit more conservative? Should they
[02:19:14] they >> I Yes. I actually I actually think it's
[02:19:17] >> I Yes. I actually I actually think it's I don't know. I don't have money in the
[02:19:18] I don't know. I don't have money in the market. I think it's a casino. Casino
[02:19:20] market. I think it's a casino. Casino pumped up by the media.
[02:19:22] pumped up by the media. >> Should they invest in the S&P 500?
[02:19:23] >> Should they invest in the S&P 500? Should they invest in Open AI?
[02:19:24] Should they invest in Open AI? Unfortunately,
[02:19:25] Unfortunately, >> oh god, no. I honestly I live in cash
[02:19:27] >> oh god, no. I honestly I live in cash right now. I live in cash. Yeah. I don't
[02:19:29] right now. I live in cash. Yeah. I don't trust the market, man. Try and
[02:19:31] trust the market, man. Try and get some gains here. I'm like I'm not
[02:19:33] get some gains here. I'm like I'm not comfortable giving financial
[02:19:35] comfortable giving financial >> advice, but it's like if you like it's
[02:19:37] >> advice, but it's like if you like it's like you're gambling.
[02:19:38] like you're gambling. >> Okay. be conservative. Things might get
[02:19:40] >> Okay. be conservative. Things might get volatile.
[02:19:40] volatile. >> Yeah, it really is. It's going to be act
[02:19:42] >> Yeah, it really is. It's going to be act as you would with volatility. Take the
[02:19:43] as you would with volatility. Take the gains when you've got them.
[02:19:45] gains when you've got them. >> Don't sell everything, but be suspicious
[02:19:48] >> Don't sell everything, but be suspicious of tech. Like, that's actually the
[02:19:49] of tech. Like, that's actually the biggest thing. It's like be suspicious
[02:19:50] biggest thing. It's like be suspicious of what they're promising. If you're
[02:19:52] of what they're promising. If you're acting based on their promises, don't
[02:19:54] acting based on their promises, don't trust the promises. Trust that they are
[02:19:57] trust the promises. Trust that they are going to say what will make the stock
[02:19:59] going to say what will make the stock run rather than what's actually
[02:20:01] run rather than what's actually happening. and that they will find every
[02:20:04] happening. and that they will find every dodgy way to make you think something is
[02:20:08] dodgy way to make you think something is happening rather than it's actually
[02:20:09] happening rather than it's actually happening. Annualized run rate, great
[02:20:11] happening. Annualized run rate, great example. Microsoft said that they had 38
[02:20:14] example. Microsoft said that they had 38 $37 billion of annualized run rate in
[02:20:16] $37 billion of annualized run rate in AI. You hear that, you go, they made 38
[02:20:19] AI. You hear that, you go, they made 38 $37 billion, right? Wow, that's so much
[02:20:22] $37 billion, right? Wow, that's so much run rate maybe month times 12. They
[02:20:24] run rate maybe month times 12. They don't even define it, but it's built to
[02:20:26] don't even define it, but it's built to manipulate. And they do that because we
[02:20:28] manipulate. And they do that because we don't have a functional SEC and we don't
[02:20:30] don't have a functional SEC and we don't have a media environment that actually
[02:20:32] have a media environment that actually where skepticism is the priority and
[02:20:35] where skepticism is the priority and where protecting the readers is
[02:20:36] where protecting the readers is necessary.
[02:20:37] necessary. >> What would they say? They would say Ed
[02:20:39] >> What would they say? They would say Ed this technology is going to be so great
[02:20:41] this technology is going to be so great and so transformative that we are
[02:20:43] and so transformative that we are investing a ton of money
[02:20:45] investing a ton of money >> um in advance of the value and utility
[02:20:49] >> um in advance of the value and utility showing up. That's what they would say,
[02:20:52] showing up. That's what they would say, >> right?
[02:20:52] >> right? >> And I've heard your rebuttal, but I just
[02:20:54] >> And I've heard your rebuttal, but I just wanted to express I think that's their
[02:20:56] wanted to express I think that's their sentiment. I'm not defending them or
[02:20:57] sentiment. I'm not defending them or anything. I'm just I'm trying to provide
[02:20:59] anything. I'm just I'm trying to provide enough like balance to we see if we can
[02:21:02] enough like balance to we see if we can dance between these these two
[02:21:03] dance between these these two perspectives.
[02:21:05] perspectives. >> And a lot of people would say that
[02:21:08] >> And a lot of people would say that there's going to be a blood bath because
[02:21:10] there's going to be a blood bath because they can't all win big in the way that
[02:21:12] they can't all win big in the way that they're kind of describing. So,
[02:21:13] they're kind of describing. So, someone's going to have to lose. And
[02:21:15] someone's going to have to lose. And >> when one of these players starts to lose
[02:21:16] >> when one of these players starts to lose big, I think it could, as you say, there
[02:21:18] big, I think it could, as you say, there could be some kind of domino effect or
[02:21:20] could be some kind of domino effect or contraction.
[02:21:20] contraction. >> Yeah. And I think the thing that people
[02:21:22] >> Yeah. And I think the thing that people want to believe is they the com bubble
[02:21:25] want to believe is they the com bubble thing. It's like it worked out
[02:21:26] thing. It's like it worked out afterwards because Amazon, Oracle, they
[02:21:29] afterwards because Amazon, Oracle, they didn't die after the com bubble. They're
[02:21:31] didn't die after the com bubble. They're actually fine. This isn't like that.
[02:21:33] actually fine. This isn't like that. They're bigger companies. They're have
[02:21:34] They're bigger companies. They're have bigger promises. And even I'm not like
[02:21:37] bigger promises. And even I'm not like Oracle I actually think could die. I RIP
[02:21:39] Oracle I actually think could die. I RIP Larry. What couldn't happen to a nastier
[02:21:42] Larry. What couldn't happen to a nastier man? They'll probably
[02:21:42] man? They'll probably >> You don't like these people, do you?
[02:21:44] >> You don't like these people, do you? >> No, I No. Again, I asked this question
[02:21:46] >> No, I No. Again, I asked this question purely because I want an answer, not
[02:21:48] purely because I want an answer, not because I agree or disagree. But um why
[02:21:50] because I agree or disagree. But um why don't you like these these people? I
[02:21:53] don't you like these these people? I don't like being misled and I don't
[02:21:55] don't like being misled and I don't think regular people like being misled
[02:21:57] think regular people like being misled either. And I really don't think that
[02:21:59] either. And I really don't think that the average person can get away with
[02:22:01] the average person can get away with bullshitting as much these companies do.
[02:22:03] bullshitting as much these companies do. And I don't think the average person
[02:22:05] And I don't think the average person gets anywhere near the level of
[02:22:06] gets anywhere near the level of affordance for failure and lying as
[02:22:08] affordance for failure and lying as these companies do. And I think there is
[02:22:10] these companies do. And I think there is a real economic and human cost to
[02:22:13] a real economic and human cost to allowing these companies to run rampant
[02:22:15] allowing these companies to run rampant and promise the world and never really
[02:22:17] and promise the world and never really get called up on it. The tepid nature of
[02:22:19] get called up on it. The tepid nature of criticism these days is so frustrating.
[02:22:22] criticism these days is so frustrating. There are some really great critics out
[02:22:23] There are some really great critics out there that really great people, but it's
[02:22:26] there that really great people, but it's like
[02:22:27] like seeing these ultra rich, ultra wealthy,
[02:22:30] seeing these ultra rich, ultra wealthy, ultra powerful people lie through their
[02:22:32] ultra powerful people lie through their teeth or misstate or whatever
[02:22:34] teeth or misstate or whatever people want to call it, it turns my
[02:22:36] people want to call it, it turns my stomach. And I hate seeing people being
[02:22:39] stomach. And I hate seeing people being misled. And I feel like I write at such
[02:22:41] misled. And I feel like I write at such length because I really want people to
[02:22:42] length because I really want people to see why I've come to a conclusion. Am I
[02:22:44] see why I've come to a conclusion. Am I right? Am I wrong? I think I am. Of
[02:22:45] right? Am I wrong? I think I am. Of course I do. But I also
[02:22:49] course I do. But I also I just find it loathome. I find these
[02:22:51] I just find it loathome. I find these companies don't make good products
[02:22:52] companies don't make good products anymore. They don't care about their
[02:22:54] anymore. They don't care about their customers and and they treat their
[02:22:57] customers and and they treat their customers with contempt.
[02:23:00] customers with contempt. >> If people want to go read more about
[02:23:01] >> If people want to go read more about your work, um you have a great Substack
[02:23:03] your work, um you have a great Substack >> Ghost actually. It looks exactly like I
[02:23:05] >> Ghost actually. It looks exactly like I moved off of Substack in 2024.
[02:23:07] moved off of Substack in 2024. >> Oh, okay. And you also have a podcast
[02:23:09] >> Oh, okay. And you also have a podcast you do.
[02:23:10] you do. >> Yeah, Better of Flame.
[02:23:11] >> Yeah, Better of Flame. >> Um I'm going to link both of them below.
[02:23:12] >> Um I'm going to link both of them below. So, if anyone wants to read more, get
[02:23:13] So, if anyone wants to read more, get more detail and and follow Ed. I think
[02:23:15] more detail and and follow Ed. I think it's
[02:23:16] it's >> I would highly recommend. It's it is
[02:23:17] >> I would highly recommend. It's it is fascinating. And you know what? One of
[02:23:19] fascinating. And you know what? One of the things people um sometimes struggle
[02:23:21] the things people um sometimes struggle with when they listen to podcasts is you
[02:23:22] with when they listen to podcasts is you get lots of different opinions. And
[02:23:24] get lots of different opinions. And weirdly, I think they think of some
[02:23:26] weirdly, I think they think of some people assume podcasts are going to be
[02:23:27] people assume podcasts are going to be like one person saying the same thing as
[02:23:29] like one person saying the same thing as the next person and then the next
[02:23:30] the next person and then the next person. That is just not the nature of
[02:23:32] person. That is just not the nature of information in the world and opinions
[02:23:33] information in the world and opinions and progress and discussion. What what
[02:23:35] and progress and discussion. What what happens is people have different
[02:23:37] happens is people have different opinions. And I think my job, but also
[02:23:38] opinions. And I think my job, but also the listener's job is to try and pass
[02:23:40] the listener's job is to try and pass through it and over time collect more of
[02:23:43] through it and over time collect more of these reference points from different
[02:23:44] these reference points from different people and and do your own research.
[02:23:47] people and and do your own research. >> Yeah. whether it's on your health or
[02:23:49] >> Yeah. whether it's on your health or whether it's on something like this is
[02:23:50] whether it's on something like this is to watch endear and research and to
[02:23:52] to watch endear and research and to learn and I would say also never believe
[02:23:55] learn and I would say also never believe one person never believe one particular
[02:23:56] one person never believe one particular perspective religiously you know collect
[02:23:59] perspective religiously you know collect a body of evidence and follow follow the
[02:24:01] a body of evidence and follow follow the evidence yourself but I love watching
[02:24:03] evidence yourself but I love watching your YouTube um because it provides a
[02:24:07] your YouTube um because it provides a different opinion and that challenges me
[02:24:10] different opinion and that challenges me to think beyond my current opinion about
[02:24:13] to think beyond my current opinion about what might be possible so when I've
[02:24:15] what might be possible so when I've heard you talking about how this is an
[02:24:17] heard you talking about how this is an economic bubble and I've heard you talk
[02:24:18] economic bubble and I've heard you talk about the capex spend on with these big
[02:24:20] about the capex spend on with these big sort of frontier AI labs. It really did
[02:24:24] sort of frontier AI labs. It really did make me pause for a second and it really
[02:24:26] make me pause for a second and it really did make me consider
[02:24:28] did make me consider that there could be a bit of fazy going
[02:24:30] that there could be a bit of fazy going on here.
[02:24:31] on here. >> Yeah.
[02:24:31] >> Yeah. >> And then it made me reflect on history
[02:24:32] >> And then it made me reflect on history and go, you know, through history
[02:24:34] and go, you know, through history there's always a bit of fazy in these
[02:24:35] there's always a bit of fazy in these moments and oh that's an interesting
[02:24:37] moments and oh that's an interesting take on what's going to happen in 2027
[02:24:38] take on what's going to happen in 2027 2028 when there's a bit of a market
[02:24:40] 2028 when there's a bit of a market pullback and so I highly recommend
[02:24:42] pullback and so I highly recommend people go watch because you do you
[02:24:43] people go watch because you do you challenge me to think differently. Um,
[02:24:44] challenge me to think differently. Um, >> yeah.
[02:24:45] >> yeah. >> And we need some of those contrarian
[02:24:46] >> And we need some of those contrarian voices to to have honest discussions.
[02:24:49] voices to to have honest discussions. So, thank you for doing what you do.
[02:24:50] So, thank you for doing what you do. Really appreciate it. And I find you to
[02:24:52] Really appreciate it. And I find you to be a very compelling, captivating
[02:24:53] be a very compelling, captivating communicator. And I've I feel like I've
[02:24:55] communicator. And I've I feel like I've learned a lot today. So, I appreciate
[02:24:57] learned a lot today. So, I appreciate that. We have a closing tradition.
[02:24:58] that. We have a closing tradition. >> Yeah.
[02:24:59] >> Yeah. >> Where the last guest leaves a question
[02:25:00] >> Where the last guest leaves a question for the next guest not knowing who
[02:25:01] for the next guest not knowing who they're leaving it for. And the question
[02:25:02] they're leaving it for. And the question left for you is given that high quality
[02:25:05] left for you is given that high quality relationships are important for health
[02:25:06] relationships are important for health and longevity, what should we be doing
[02:25:08] and longevity, what should we be doing to improve our relationships and social
[02:25:11] to improve our relationships and social connection? So this is actually
[02:25:14] connection? So this is actually connected to the AI bubble. So I am a
[02:25:17] connected to the AI bubble. So I am a critic. I'm a skeptic. What quote I have
[02:25:20] critic. I'm a skeptic. What quote I have found that showing and appreciating and
[02:25:24] found that showing and appreciating and loving the people around you and
[02:25:26] loving the people around you and uplifting them and me and and raising
[02:25:28] uplifting them and me and and raising them up as you succeed is the way we do
[02:25:30] them up as you succeed is the way we do that. Your success should be everyone
[02:25:31] that. Your success should be everyone around you. It's not economic. It's
[02:25:33] around you. It's not economic. It's talking about Matt Hughes for a while
[02:25:35] talking about Matt Hughes for a while made me really happy. This whole thing
[02:25:37] made me really happy. This whole thing has been at times quite grueling and
[02:25:39] has been at times quite grueling and quite negative and quite brutal. But the
[02:25:42] quite negative and quite brutal. But the love I found and the joy I found from
[02:25:45] love I found and the joy I found from community and the people around because
[02:25:46] community and the people around because even in the in the small groups of
[02:25:49] even in the in the small groups of haters even like Gary Marcus and sort of
[02:25:50] haters even like Gary Marcus and sort of the people I talked to Edward on Grao
[02:25:52] the people I talked to Edward on Grao Jr. Molly White, Brian Merchant, there
[02:25:55] Jr. Molly White, Brian Merchant, there are so many people who have been loving
[02:25:56] are so many people who have been loving and caring. And I think within
[02:25:59] and caring. And I think within especially these very critical moments
[02:26:01] especially these very critical moments when you're like very much dialing in on
[02:26:03] when you're like very much dialing in on how negative things are, how bad things
[02:26:05] how negative things are, how bad things are, finding the people who maybe find
[02:26:08] are, finding the people who maybe find it repulsive, too. Finding the people,
[02:26:10] it repulsive, too. Finding the people, >> finding your people who can be and the
[02:26:12] >> finding your people who can be and the people who will talk to you about it.
[02:26:14] people who will talk to you about it. Even like Troy and Jake, my my trainers
[02:26:16] Even like Troy and Jake, my my trainers who's so excited about this. um even
[02:26:18] who's so excited about this. um even talking to them about the as normal
[02:26:20] talking to them about the as normal people knowing that there are people
[02:26:22] people knowing that there are people there going through their own struggles
[02:26:23] there going through their own struggles but also to just give you the
[02:26:25] but also to just give you the perspective and also remind you that you
[02:26:28] perspective and also remind you that you are human to and focus I know this is
[02:26:29] are human to and focus I know this is kind of a all over the place point but
[02:26:31] kind of a all over the place point but it's just it's really easy to get hard
[02:26:33] it's just it's really easy to get hard locked on everything in life and to
[02:26:35] locked on everything in life and to >> kind of get away from why you do things
[02:26:37] >> kind of get away from why you do things and focus too much on the work when the
[02:26:39] and focus too much on the work when the most important thing at times is just to
[02:26:41] most important thing at times is just to know there are other people feeling the
[02:26:43] know there are other people feeling the way you do and when I hear from my
[02:26:44] way you do and when I hear from my listeners and my readers a lot the most
[02:26:46] listeners and my readers a lot the most common thing they feel is they feel like
[02:26:48] common thing they feel is they feel like they have a voice and they feel like
[02:26:49] they have a voice and they feel like someone is there for you.
[02:26:50] someone is there for you. >> And I don't think it can be understated
[02:26:53] >> And I don't think it can be understated how much it means when you just reach
[02:26:54] how much it means when you just reach out to someone you love and tell them
[02:26:55] out to someone you love and tell them you love them. Tell them their
[02:26:57] you love them. Tell them their rocks. Say that their bangs. Tell
[02:26:59] rocks. Say that their bangs. Tell everyone you when you like an artist or
[02:27:01] everyone you when you like an artist or a writer they were a podcast like this.
[02:27:03] a writer they were a podcast like this. Tell them you love it. We don't
[02:27:05] Tell them you love it. We don't do this enough and we need to do it
[02:27:07] do this enough and we need to do it more. Well, that's a good closing
[02:27:09] more. Well, that's a good closing message. So, if you do have you have
[02:27:10] message. So, if you do have you have enjoyed the conversation today with Ed,
[02:27:12] enjoyed the conversation today with Ed, please do let Ed know that you love it
[02:27:13] please do let Ed know that you love it down below. Um, but please do leave your
[02:27:15] down below. Um, but please do leave your opinions down below and I shall read all
[02:27:17] opinions down below and I shall read all of them. Ed, thank you so much. I'll
[02:27:18] of them. Ed, thank you so much. I'll link to your website, but also to your
[02:27:21] link to your website, but also to your YouTube channel where people can learn
[02:27:22] YouTube channel where people can learn more and I would highly recommend you do
[02:27:24] more and I would highly recommend you do because it is truly fascinating and I
[02:27:25] because it is truly fascinating and I think we need more voices that are
[02:27:27] think we need more voices that are demystifying a lot of the fugazi and the
[02:27:28] demystifying a lot of the fugazi and the narrative in this moment in time and you
[02:27:30] narrative in this moment in time and you are certainly one of them. I really
[02:27:31] are certainly one of them. I really enjoyed the conversation. Thank you so
[02:27:32] enjoyed the conversation. Thank you so much.
[02:27:33] much. >> YouTube have this new crazy algorithm
[02:27:34] >> YouTube have this new crazy algorithm where they know exactly what video you
[02:27:36] where they know exactly what video you would like to watch next based on AI and
[02:27:39] would like to watch next based on AI and all of your viewing behavior. And the
[02:27:41] all of your viewing behavior. And the algorithm says that this video is the
[02:27:44] algorithm says that this video is the perfect video for you. It's different
[02:27:45] perfect video for you. It's different for everybody looking right now. Check
[02:27:47] for everybody looking right now. Check this video out and I bet you you might
[02:27:49] this video out and I bet you you might love it.

All frames

Total: 80. Hero frames flagged with star.

• frames/frame_0001.jpg (t=00:00) frames/frame_0002.jpg (t=00:02) frames/frame_0003.jpg (t=00:03) frames/frame_0004.jpg (t=00:04) frames/frame_0005.jpg (t=00:06) frames/frame_0006.jpg (t=00:07) frames/frame_0007.jpg (t=00:08) frames/frame_0008.jpg (t=00:10) frames/frame_0009.jpg (t=00:12) frames/frame_0010.jpg (t=00:12) frames/frame_0011.jpg (t=00:12) frames/frame_0012.jpg (t=00:15) frames/frame_0013.jpg (t=00:15) frames/frame_0014.jpg (t=00:16) frames/frame_0015.jpg (t=00:18) frames/frame_0016.jpg (t=00:20)
• frames/frame_0017.jpg (t=00:21) frames/frame_0018.jpg (t=00:23) frames/frame_0019.jpg (t=00:25) frames/frame_0020.jpg (t=00:30) frames/frame_0021.jpg (t=00:32) frames/frame_0022.jpg (t=00:35) frames/frame_0023.jpg (t=00:37) frames/frame_0024.jpg (t=00:38) frames/frame_0025.jpg (t=00:41) frames/frame_0026.jpg (t=00:42) frames/frame_0027.jpg (t=00:43) frames/frame_0028.jpg (t=00:43) frames/frame_0029.jpg (t=00:43) frames/frame_0030.jpg (t=00:45) frames/frame_0031.jpg (t=00:46) frames/frame_0032.jpg (t=00:47)
• frames/frame_0033.jpg (t=00:48) frames/frame_0034.jpg (t=00:52) frames/frame_0035.jpg (t=00:55) frames/frame_0036.jpg (t=00:56) frames/frame_0037.jpg (t=00:56) frames/frame_0038.jpg (t=00:57) frames/frame_0039.jpg (t=00:58) frames/frame_0040.jpg (t=01:02) frames/frame_0041.jpg (t=01:02) frames/frame_0042.jpg (t=01:04) frames/frame_0043.jpg (t=01:07) frames/frame_0044.jpg (t=01:09) frames/frame_0045.jpg (t=01:12) frames/frame_0046.jpg (t=01:14) frames/frame_0047.jpg (t=01:16) frames/frame_0048.jpg (t=01:18)
• frames/frame_0049.jpg (t=01:21) frames/frame_0050.jpg (t=01:22) frames/frame_0051.jpg (t=01:25) frames/frame_0052.jpg (t=01:27) frames/frame_0053.jpg (t=01:30) frames/frame_0054.jpg (t=01:32) frames/frame_0055.jpg (t=01:34) frames/frame_0056.jpg (t=01:35) frames/frame_0057.jpg (t=01:37) frames/frame_0058.jpg (t=01:38) frames/frame_0059.jpg (t=01:46) frames/frame_0060.jpg (t=01:51) frames/frame_0061.jpg (t=02:09) frames/frame_0062.jpg (t=02:13) frames/frame_0063.jpg (t=02:20) frames/frame_0064.jpg (t=02:24) frames/frame_0065.jpg (t=02:36) frames/frame_0066.jpg (t=02:53) frames/frame_0067.jpg (t=02:57) frames/frame_0068.jpg (t=03:17) frames/frame_0069.jpg (t=03:19) frames/frame_0070.jpg (t=03:37) frames/frame_0071.jpg (t=03:45) frames/frame_0072.jpg (t=03:54) frames/frame_0073.jpg (t=04:01) frames/frame_0074.jpg (t=04:04) frames/frame_0075.jpg (t=04:09) frames/frame_0076.jpg (t=04:27) frames/frame_0077.jpg (t=04:29) frames/frame_0078.jpg (t=04:37) frames/frame_0079.jpg (t=04:38)
• frames/frame_0080.jpg (t=04:58)