How The AI Bet Pays Off + AI Lab Strategy Game — With David Cahn

Summary

---

Big Technology Podcast — Episode Summary

Episode Title: How The AI Bet Pays Off + AI Lab Strategy Game

Host: Alex Kantrowitz
Guest: David Cahn, Partner at Sequoia Capital
Release Date: August 5, 2026

---

Overview

Alex Kantrowitz hosts Sequoia partner and noted AI-market analyst David Cahn for an in-depth discussion about the financial and strategic realities underpinning the AI revolution. The episode tackles questions of AI’s return on investment (ROI), the feasibility of massive capital expenditures, and how major tech players are navigating what Cahn calls “the greatest strategy game in history.” The session finishes with an unexpected exploration of AI’s potential impact on human spirituality.

---

Key Discussion Points & Insights

1. The Scale and ROI Question in AI (01:30–13:13)

• The Exponential Boom in AI Investment

  • The conversation kicks off with a look back at Cahn’s writing since 2024, when he first quantified how much revenue it would take for AI investments (mostly in GPUs and datacenters) to pay off. The numbers have exploded from a “$200 billion question” to cumulative investments now in the $3–4 trillion range.
    • "For every $1 spent on a GPU, roughly $1 needs to be spent on energy costs to run the GPU in a data center. So if Nvidia sells $50 billion in run rate GPU revenue...that implies approximately $100 billion in cumulative data center expenditures." —Alex Kantrowitz [02:15]
  • Market scale comparison: Annual combined revenue for all SaaS and cloud companies hovers around $1 trillion—a fraction of what’s needed.
    • "You're talking about very big numbers here. The cloud software industry...about a $500 billion market [cloud infra], SaaS apps, another ~$500 billion. Let's call it $1 trillion total. And yet we need $4 trillion in lifetime revenue to pay back today's AI capex." —David Cahn [07:45]

• The ROI Dilemma

  • While AI’s long-term addressable market is technically all human cognitive labor (orders of magnitude larger than software itself), the timeline for payback is deeply uncertain. Investors are betting "now" for returns that may arrive decades from now.
    • "In the long run, there's no ROI question. However, there's always a timing tension in financial markets because the long run is not tomorrow. Companies are spending today." —David Cahn [09:24]

• Progress vs. Expenditure

  • "All in" behavior has led costs to scale just as quickly as revenues—fueling a cycle where both sides keep escalating.
    • "The cost side of the equation has scaled maybe as fast as the revenue side of the equation." —David Cahn [11:46]

2. Is AGI the Only Justification? (13:13–21:12)

• AGI as the Endgame

  • David notes that only the arrival of artificial general intelligence (AGI) could ever justify the astronomical investments.
    • "Nothing [short of] AGI will be enough to justify the investments now being proposed for the coming decade." —David Cahn [13:29]
  • Market moods swing between belief in a near-term payoff and anxiety about durations/match. Despite enormous near-term costs and negative free cash flow territory, companies continue to escalate their bets, driven by the belief—or at least hope—that AGI will pay off.

• Mismatched Timelines

  • Leaders like Sam Altman, Dario Amodei, and Ilya Sutskever have varied projections, with Amodei predicting sooner, others predicting later. The “wait for AGI” hypothesis still drives decision-making even as lab leaders temper timelines.
    • "A lot of the pronouncements we’re getting from the lab leaders have been sort of downplaying or slow playing, you know, that timeline." —Alex Kantrowitz [16:21]
    • "AGI is going to be amazing, right? Like, in 50—when I’m 80 years old—the world is going to be completely different...But the timeline and the path dependency, how we get there, matters." —David Cahn [18:46]

3. The Bubble Debate (21:12–24:28)

• Why "Bubble" is a Loaded, Unhelpful Term
  • Both guests agree the “bubble/no bubble” discourse obscures more than it reveals. Financial incentives shape public postures, and actual outcomes are much more nuanced.
    • "The word bubble has all of this emotional baggage associated with it. ...People almost stop thinking. You say the word bubble, and their brain shuts down." —David Cahn [21:54]
  • Instead, the real focus should be on navigating cycles and surviving interim corrections, not obsessing over whether present conditions are a “bubble.”
    • "In the long run, this technology is obviously going to work. We need to make sure that we’re the winners." —David Cahn [23:54]

---

Company-by-Company: The AI Lab Strategy Game (28:07–61:12)

David Cahn offers chess- and strategy-game-inspired reflections on how each of the key tech giants are playing the "AI game" and what resources or philosophies define those games.

Framing

• StarCraft & Azad Metaphors: AI as a resource-allocation game where each participant’s intrinsic capabilities shape the best strategy.
  • "What are your resources and how do you allocate those resources? That’s the most important question facing every big tech company." —David Cahn [28:07]

Anthropic (29:39–33:07)

• Strategy: Corner the world’s talent in AI and go all-in on AGI.
  • "The strategy is: corner the world’s talent in AI and win, period...Talent is the scarce resource right now." —David Cahn [29:57]
• Incidents like Meta’s 2025 talent raid failed to dislodge Anthropic’s core team, evidence of their strong culture and focused mission.
• Weakness: If AGI becomes commoditized, being “best at frontier” could lose value—but you must pick a horse and drive it.

OpenAI (35:20–37:44)

• Strategy: Out-execute everyone else in racing towards AGI, betting the timing will reward high risk/aggression.
  • "OpenAI’s strategy is: everybody underestimates AGI, all in on AGI, we’re going to be the most aggressive...and win." —David Cahn [35:30]
  • OpenAI’s success now seen both in technology and in the ability to aggregate compute/talent at all costs.
  • Weakness: Most risk if revenue doesn't appear in time to match costs.

Google (37:44–45:10)

• Strategy: Unclear, but inclined toward building everything in-house (TPUs, models), internally vertical and insular.
  • "Google is hard because it doesn't seem to be run by a single individual with a single vision." —David Cahn [38:14]
  • Strong technical assets (especially TPU) but lacks cohesive, founder-style execution.
  • Possible alternative: Spin out TPU as a product/ecosystem business, compete directly with NVIDIA.
  • Weakness: Search and ad-driven revenue streams increasingly vulnerable to AI-driven interface changes.

Meta (48:00–51:10)

• Strategy: Buy/hoard top AI talent, build a "mercenary army," and wait for the emergence of a winning consumer AI product to copy and distribute.
  • "Buy talent, it’s a mercenary army. Can a mercenary army do as well as a missionary army?" —David Cahn [48:11]
  • Faces challenge in weak organic talent flows and possible organizational dysfunction.
  • Ultimate bet: That succeeding in talent aggregation will let them catch/ride the AGI wave.

Microsoft (51:10–53:58)

• Strategy: Play both sides—own a large piece of OpenAI, cultivate distribution to enterprise, and stand to win either way (commoditized AI or frontier models).
  • "Whoever wins will win, you know, and I love that." —David Cahn [51:50]
  • Satya Nadella lauded for "chess master" dealmaking and strategic patience.
  • Possibly underutilizing M&A muscle but remains adaptable and conservative.

Amazon (53:58–57:44)

• Strategy: Do “very little” in terms of original AI lab efforts, but invest heavily in cloud/data center scaling. Neutral, defensive posture—"steady as she goes."
  • "Do very little, sell compute, profit." —Alex Kantrowitz [54:03]
  • At risk of losing edge to Azure/Google in AI; “resource curse” of immense capex and committee-driven inertia.

Nvidia (57:44–61:12)

• Strategy: Make the AI pie as big as possible, enable the entire ecosystem, take a small cut everywhere, and avoid excessive price gouging or ecosystem exclusion.
  • "Nvidia: prop up the AI ecosystem... If AI works and doesn’t crash, Nvidia is going to be in a great spot." —David Cahn [57:49]
  • Jensen Huang’s partnership and ecosystem-first philosophy credited with historic success.

SpaceX/Elon Musk (61:12–63:39)

• Strategy: Civilization-scale mission focus—move humanity forward, aggregation of top talent, and applying AI strategically to largest possible problems.
  • "To me, SpaceX is like a meta commentary on financial markets...I don’t think Elon is fundamentally dollar motivated. I think Elon is fundamentally mission motivated." —David Cahn [61:38]

---

Notable Quotes & Memorable Moments

• "In the long run, there’s no ROI question. However, there’s always a timing tension in financial markets because the long run is not tomorrow." —David Cahn [09:24]
• "These people [company leaders] are grandmasters playing 4D chess for the highest stakes in the universe." —David Cahn [29:24]
• "Google has amazing resources...but when you think about the gameplay, it does sometimes feel disjointed." —David Cahn [38:14]
• "Buy talent, it’s a mercenary army...Can a mercenary army do as well as a missionary army?" —David Cahn [48:11]
• "Apple is the one with the do-nothing strategy...You can imagine a scenario where Apple looks great in 10, 20 years because of the decisions they made." —David Cahn [54:53]

---

AI, Spirituality, and the Human Quest for Transcendence (63:39–70:25)

• Does AI Fulfill a Human Spiritual Need?

  • David shares how the quest for “transcendence,” a role historically filled by religion, may now be projecting itself onto AI and its creators.
    • "Religion in my mind is this…thousands of years quest to understand the universe. …These people are trying to create a new God and maybe that will reinfuse the world with some spirituality." —David Cahn [64:41]
  • Silicon Valley’s godlessness, loss of community, and the rise of therapy culture parallel, and perhaps motivate, both the AI quest and spiritual searching.
  • "This AI quest is sort of fulfilling this need that people have for transcendence. And I wonder if that is part of what’s powering it, why it’s become so big." —David Cahn [68:20]

• Will Humans Worship AI?

  • The question is open, but Cahn admits he’s only in “inning one or two” of contemplating these issues. He hopes that as physical toil is automated, people will achieve richer lives, but the integration of work and spiritual “Torah” is essential.
    • "There’s a Jewish phrase which is, there’s no work without Torah and there’s no Torah without work… you need the spirituality quest and you need to be grounded by some real things." —David Cahn [70:12]

---

Timestamps for Major Segments

• 01:30–13:13: The size and ROI of the global AI investment
• 13:13–21:12: AGI as the only possible ROI and the long/short-term tension
• 21:12–24:28: The bubble debate and why it’s unproductive
• 28:07–61:12: AI Lab Strategy Game – company-by-company breakdown
• 63:39–70:25: AI as a spiritual phenomenon and society’s search for transcendence

---

Final Takeaway

This episode presents a rare, nuanced look at the extraordinary gamble of the AI era: a bet so big that only a revolution on the scale of AGI could ever justify it. The strategic positioning of tech giants is less a business arms race than a set of grandmasters playing for existential stakes—each defined as much by their internal cultures, philosophies, and leader’s personalities as by raw resources. Ultimately, the episode leaves open the possibility that AI’s true legacy may be cultural or spiritual, answering humanity’s perennial need for transcendence as much as the market’s demand for ROI.

---

Transcript

[A]

How much money does the AI industry need to generate to pay back its investments? And which company is employing the winning strategy? That's coming up with Sequoia partner David

[B]

Kahn right after this Insurance isn't one size fits all, and shopping for it shouldn't feel like squeezing into something that just doesn't fit. That's why drivers have enjoyed Progressive's name your price tool for years. With the name your price tool, you tell them what you want to pay and they show you options that fit your budget enough. Hunting for discounts, trying to calculate rates, and tinkering with coverages. Maybe you're picking out your very first policy, or maybe you're just looking for something that works better for you and your family. Either way, they make it simple to see your options. No guesswork, no surprises. Ready to see how easy and fun shopping for car insurance can be? Visit progressive.com and give the name your price tool a try. Take the stress out of shopping and find coverage that fits your life on your terms. Progressive Casualty Insurance Company and Affiliates Price and Coverage Match limited by State Law

[A]

welcome to Big Technology Podcast, a show for cool headed and nuanced conversation of the tech world and beyond. We have a great show for you today. David Kahn, who's writing we've read on the show week after week, is here with us today to talk about what it's going to take to pay back all the investment going into AI. And then we'll go company by company and decide which ones have the winning strategy and which ones might not. So, David, it's great to see you finally. Welcome to the show.

[B]

Thanks for having me, Alex.

[A]

So your writing came onto our radar first when you were basically talking in 2024 about the ROI needed to make the AI bets pay off. This was long before the term AI ROI was in vogue and there was a sort of determination that there was some promise in this technology. And so of course you'd want to spend a lot of money in it because you don't want to be left out, right? But you actually very early on started to say, hey, let's at least put some numbers around this to see what it will take to pay off what was coming in. And it started in 2024 where you wrote this AI's $200 billion question. And for us it caught our attention because, wow, like it. Basically you demonstrated that AI would have to make Lifetime $200 billion to pay off its investments, and that was starting to look like real money. So let me read a quick selection from that and we'll kind of talk about how it might apply today. So you said for every $1 spent on a GPU, roughly $1 needs to be spent on energy costs to run the GPU in a data center. So If Nvidia sells $50 billion in run rate GPU revenue by the end of the year, that implies an approximately $100 billion in cumulative data center expenditures. Let's assume those building them need to earn a 50 applies for each year of current GPU capex. 200 billion of lifetime revenue would need to be generated by these GPUs to pay back the upfront capital investment. So the total number in 2024 just a mere two years ago was $200 billion lifetime. Let me talk about where we are today. So conservatively right now we're looking at $2 trillion in capex in big tech capex this year. Next between cumulative together 2026, 2027 projected it's more than than 2 trillion. But I'll just say 2 trillion to be conservative about it. So by your math that would be 4 trillion in lifetime revenue on this AI infrastructure needed to make that money back. So in terms of tech revenue, give us a scale, like a sense of scale of how much is 4 trillion? Really? Is this something that tech companies make regularly? They don't make often. And how feasible is it for this current level of investment to earn that money to pay back lifetime and make those investors whole?

[B]

Well, maybe first a couple of comments that. First the 200 billion was 2023, 600 billion 24 anyways, regardless, these numbers have gotten really big. It's funny to almost hear the numbers from then because they feel quaint. Things have gotten obviously mega sized since then and had a big reinflection in 2026 which is why you had $200 million question and 600. So a 3x growth 2025 it was 850 so it had slowed down and then it more than it about doubled in 2026 to 1.5 trillion. And then as you say, if you look at 2027 and the forecast there, it's obviously going to scale past 1.5 trillion at this point. And then the last thing to say there is these numbers are cumulative. So if you really want to ask yourself hey, what's the total capex burden? You say for every dollar capex we eventually need to get an roi, how much? How do you actually have to add up all of those numbers? So it's 200 plus 600 plus 850 plus 1.5. You basically have about $3 trillion that needs to get paid back just since ChatGPT. And then as you say, if you add 2027, it's going to get larger. And yes, I think to your point, I think we're all starting to recognize how big these numbers have gotten. The first post that kind of went Viral was the $600 billion question. That was summer of 2024. And I think the reason for that at that point, if you remember that was that was when Nvidia became the most valuable company in the world. And so when I first started publishing these posts, no one really cared. It wasn't really a topic of conversation that people are focused on. And what's crazy to me is that two and a half, you know, two and a half years later from the original post, this is still the number one topic of conversation. And so anyways, hopefully we'll take through it quickly. I think probably your listeners have heard a lot about this topic in the last year. And the ROI debate, which is now what everyone calls this, is raging full speed. And every time a big tech company reports earnings, you hear the debate on both sides. And so my original intent was to ask the question. I think it's important to look at both sides. As you said in your intro, we try to be cool headed and just look at both sides of these things. On the one hand, AI is going to be probably the greatest revolution in human history. Today humans do 99% of cognitive work. In 50 years humans are going to do maybe 1% of cognitive work, right? So there's a huge revolution that's coming and we all see that. And yet on the other hand, wearing our financial analyst hats, looking at markets, things go through boom bus cycles, we have to be cool headed about the timeline that it takes to get there. Where the investment goes, who gets the returns. And one thing that we've learned from studying history is there are winners and losers. And so in these moments of immense hype and immense excitement, there's sort of this assumption that everybody's going to be a winner and that at some point there's an assumption that everyone's going to be a loser. And in reality, and this is what we'll get into in some of the game theory and some of the company specific stuff, some people are going to be winners and some people are going to be losers. And I think maybe the interesting conversation for those of us who are not in the arena and are looking in is, well, who's going to win? Right? That's, I think as humans we're just fascinated by that question. And I'm fascinated by that question. And I think the closer you get and the deeper you get, the more it becomes apparent that it's these human personalities. It's not some abstract math, it's actually human personalities driving it. And so it all started with a $600 billion question. But I think downstream of that is humans and company cultures and ecosystem effects as these companies interact with each other.

[A]

Okay, yes, that's true, and we're going to get into that side of things. But I'm going to go back to the initial question, which is the sense of scale of that type of revenue. Why don't you compare it to like what we see in SAT? Like the total, I think the total SaaS industry makes less than a, less than a trillion a year. All big tech makes less than a trillion a year. Right. So just give us again like a sense of scale of what's going to be needed to return this 4 trillion and whether you think it's feasible. Yeah.

[B]

In one of the first follow up posts in summer 24, I did a post in the game theory of AI capex and I sort of tried to quantify what are we talking about here. And to your point, you know, let's just use round numbers here and say the cloud software industry has two big buckets. There's cloud infrastructure, AWS, Azure, GCP, about a $500 billion market. And then you have SaaS applications, roughly a $500 billion market. Let's call it a trillion total. It's actually less than that, but let's call it a trillion total. And so you're talking about very big numbers here. And then I think the second order thing that people talk about in AI investing is okay, fine, but we're not actually. The TAM that AI addresses is not software revenue. And I think that's correct. The TAM is actually human labor. And so human labor is a much bigger addressable market. While I think that's true, somebody sent me a stat recently that was like, you know, there's x trillion dollars of services revenue and if AI automates 10% of that and the labs capture all of that value, then we gotta pay back. And it's like, okay, that's great, but that's a huge assumption to be making, which is you need to have 10% of all of the services industries get automated and then there's some question of value capture downstream of that. And so I do think in the long run, like I said, I think 99% of cognitive labor is going to be done by AI. So in the long run there's no ROI question. However, there's this tension, there's always a timing tension in financial markets because the long run is not tomorrow. Companies are spending today. And I think there's this question of, and there's this constant back and forth in financial markets with people trying to figure out like when is the long run coming and are we going to get the payback in the short term?

[A]

Yeah. So I guess your answer here is it is feasible, but the question is does it come on the timetable that is going to be acceptable? And sort of the interesting thing that we're going to find out is whether there is that sort of duration mismatch

[B]

and not only is it feasible, but we're making tremendous progress. Right. I think it is worth saying, and I said this in the 1.5 trillion post, I mean what has happened with Anthropic in the last year is nothing short of mind blowing to anyone who studied companies. Right. It's the fastest growing company in history. And so I think it's more than feasible. There's real tangible progress. When I trace back when I first published the $600 billion question, I said basically OpenAI is the Lion's share of revenue. I think at that time it was 12 billion revenue. Today it's 100 billion plus across OpenAI Anthropic. Now there's still two companies driving the vast, vast, vast majority of the revenue in AI. And so there's still a long way to go here. But there has been tremendous progress. And so I think that's, that's, that's important and I think that's great.

[A]

Yeah, I definitely thought about that when I was going back and reading some of those old posts that you wrote. You're sort of tallying up like let's get the feasible number that you know, we might see in terms of AI revenue. And you had OpenAI there and Anthropic there and then all the cloud services. And one of the interesting things about this technology is, you know, it has sort of given birth to brand new use cases and new very lucrative lines of business. Like I'll go back to one of your 2025 posts talking about the ChatGPT and Anthropic revenue. ChatGPT has continued its epic rise north of 12 billion run rate in reven and Anthropic has reached 5 billion in run rate revenue. And there's new club of companies quickly scaling from 0 to 100 million in revenue. You know, I think the only person that was really convinced that that anthropic revenue was going to 10x this year, which is clearly what's going to happen, was Dario, who like said it very clearly that they went from 1 billion or they went from. Yeah, they went 1 billion in 2024, they were going to do 10 billion in 2025 and who knows what happens? And it seemed crazy at the time, but because of what this technology enables, cloud code and cloud cowork showed up and then Anthropic went from having a decent sized AI API business to a powerhouse product business.

[B]

Nothing would make me happier than to see all these questions get answered. And so I think for me it's always been, let's just look at both sides of the equation. The reality is the cost side of the equation has scaled maybe as fast as the revenue side of the equation. Right. And so I think you should have this numerator and denominator and you sort of make progress on one side, but then you sort of have more of a hole to fill. And I think one thing we've seen in 2026 in particular is that on the back of all of the AI coding, scaling hyperscalers double down on capex. And so I think when you look at this year in particular, you know, is moving so fast that I feel like we sometimes it's like people can't remember what happened less than two weeks ago. But if you rewind back to January 2025, which feels like forever ago, Microsoft and Amazon were pulling back on capex and there was this moment of like, well, we'll have Oracle do it and it's super risky. Why not have Oracle do it? And then Microsoft and Amazon both got penalized by the stock market where people said, oh, they're not bullish enough on AI. And so then they had to come back to the game, if you will. And now we've reached a point in 2026 where not only is everybody in the game, but everyone is massively accelerating. And there's no, you know, I think we were, at some point there was some question of like, are people going to try to rationalize things? Are people going to look at the math? I think we reached a point where people just don't care about the math anymore at all. It's all in, all the way, let's see what happens. And that's how I view 2026. And I think there was a real catalyst for that earlier this year.

[A]

Okay, so speaking of not caring about the math and going all in, one thing that you wrote Recently, I think, was in 2025. Also, you wrote, One thing has become clear. Nothing. AGI will be enough to justify the investments now being proposed for the coming decade. Do you still believe that?

[B]

Yeah, I do. I think this when you and I first started exchanging emails on this topic. It's funny, actually, I remember writing that post, I just read Hyperion, which is the name of Meta's new data center. So there's something to that. But at that time, again, to remind the audience, at that time everyone was talking about 10 gigawatts, then 30 gigawatts, then 100 gigawatts of CapEx. And so, and those are just astronomically large numbers, far beyond where we are today, to be clear. And so if you kind of think about Those dollars of CapEx, the only possible way to pay those dollars back is going to be AGI. And so what I think the market kind of gets wrong and is just inherent to Wall Street. But Wall street sort of always has this view of like, is the stock market going to go up 2% or is it going to go down 2%? Right? It's like this, just constant volatility. And everyone, you know, if the market's down 10% in a month, it's like, oh my God. If the market's up 10% in a month, oh my God. In reality, I think we are reaching this, like sort of bifurcated path where path one is like, we get AGI, we pay back all these numbers and more. It's the greatest technology in human history, all of the things that we hear in the media all the time. And then path two is we got the timing wrong. There's no next application after coding and you have a big reckoning that has to come. And I think in some ways, I think the market underestimates the probability of AGI and underestimates the probability of some correction and massively overestimates the status quo bias, which is the current status quo of we're spending $200 billion a year on CapEx. And yes, we're renting our GPUs. So revenue is accelerating, but fundamentally there's not a new business line coming out of it that is not sustainable. And so I do think, and if, for what it's worth, this is why I talk about kind of the game theory and logic of the people leading these companies. You look at Sundar, you look at Satya, you look at Sam, you look at Elon, you look at Dario. They've told us what they believe. They believe that they're chasing AGI. It's very clear. I think there is no debate about that. These guys have been extremely clear about what they're doing. The financial market kind of likes to ignore sometimes what people are saying, because kind of hard to own Google on the assumption of AGI. It's just like hard for our brains to grok that. And so we sort of pare that back into some, oh, there's going to be some roi. You know, Microsoft's going to announce earnings, they're going to announce Azure incrementality of x or x plus 2%. And we're going to. The stock's going to move on this, like, random number that doesn't really matter in reality. What matters is like, are we going to get to AGI or not? And are we going to have enough evidence that they can continue to spend at this scale when now we're getting into negative free cash flow territory. We're getting to a scale where it is going to be harder and harder to keep spending at this scale. But all the evidence suggests that the hyperscalers are going to continue aggressively moving in this direction.

[A]

You know, what's interesting is that it seems like right now a lot of the market and the investment is going into these companies sort of imagining that AGI is going to happen pretty soon. But as you've pointed out, uh, a lot of the pronouncements we're getting from the lab leaders have been sort of downplaying or slow playing, you know, that time, that timeline. Uh, Dario, of course thinks it might happen, you know, this year, next year. Um, but Sam Altman talked about a gentle singularity. Ilya talked about how, like, pre training is over. Yann Lecun has been like, on the warpath talking about how LLMs are not the route to AGI. So whatever you make of that, that mismatch between, like, what we're hearing from the people who are building this technology and what the market and the investment seems to be pointing to, which is like, if I was thinking of places where this is out of sync, that might be like the number one place.

[B]

Yeah, this is the number one thing that I was writing about in 2025. I was just obsessed with this topic in 2025. Yeah. Part of the reason why I was so obsessed with it. You know, I've been investing in AI for about 10 years now, so I'm in a massive AI bull for a very, very long time. And, and I started investing in AI a year after the Transformer paper. So early, but not you know, there are people who are earlier, Right. And I always try to follow. My philosophy is kind of as an investor, your job is to follow the really smart entrepreneurs and eventually things become consensus because the evidence is so obvious. And your job as an investor is just to be ahead of that consensus. And so maybe you start investing in AI five years before ChatGPT, but you're not going to start investing 10 years before ChatGPT. You have to follow the. And what I started to see is that Ilya and Sam and Greg and all these guys are coming out and they're saying it's AGI is 10 years away. And by the way, as someone who's been following AI for 10 years, that makes a lot of sense. These things take time. And in 2024 and 2025, the thing that I had gone deep on, I published this piece, Server Steal and Power, and I got extremely deep on just like, how do you actually build a data center? I'd flown, I'd visited a bunch of data centers, and I sort of had this realization that, well, it takes two years from the time that you announce you're doing a data center, the data center getting built, maybe it's going to be longer because of all of the community pushback and stuff that we're seeing right now. And so just this idea, I think in an Excel spreadsheet or in a chart, it's very easy to show exponential scaling. But then in the physical world, exponential scaling is very challenging. And so I had this view, and again, I was sort of obsessed with this in 2025 because I thought there was such a divergence between the way the markets were talking about this and the way the really smart technical people were talking about this, where AGI could be 10 or 20 years away. And then the thing that I've been saying, and I think I've been saying this on Repeat maybe for two years now, but it's like, AGI is going to be amazing, right? Like in 50, when I'm 80 years old. It's something I said a lot of times, like when I'm 80 years old, the world is going to be completely different. Everything about the way we live is going to be different. So in some ways, all of the optimistic forecasts are right. And if anything, they're underestimating how much the world is going to change. But the timeline and the path dependency, how we get there matters. Financial markets care about path dependency. And I think a lot of, you know, that's where the winners and losers come out, because on the path to get there, you have to make the right strategic decisions. And if you're a grandmaster, which I think Elon is and Sam Altman is, and some of these guys are grand masters, they're trying to plan out, like, okay, well, if this happens, then this happens, then this happens. And I think that the leaders of these labs are planning many, many moves ahead. And we're oversimplifying. When we just look at the financial statements we have to look at what's the master plan and is it going to work, and whose master plan kind of makes the most sense. And that's what I'm trying to collect data on. In one of the posts, I said, like, every time I see an AI headline, I think like, you know, Knight moves to E6 or something, right? Like, what is this sort of move? What does this imply about the chessboard? And I'm kind of building this world model, if you will. Like, I have this sort of a master world model. It's not correct, obviously. It's incorrect in 100 ways. But I'm constantly updating this, this master world model with every decision. Some decisions are consistent, so it's a very small update. And some decisions are new and they actually require some updating of, like, what is actually going on, what is the underlying strategy? And if you can hold that world model in your brain, that actually enables you to forecast what's going to happen in five or six years. Far more so than, like, Azure incrementality is X versus Y. And to me, one of the biggest updates was last year when all these guys came out and said the same thing, which is AGI is 10 years away. That was a huge update.

[A]

Okay. Yeah, I definitely want to get into sort of the strategic moves. I've teased it a couple of times. I'm going to ask you one more question on this section, then we're going to take a break and move on. You know, one of the things that, that I wanted to bring up to you was, and you've talked about this a lot is this notion of an AI bubble. And you know, I'm not going to ask you whether we're in an AI bubble or not. Like you said before that there are elements of this. That is. But I guess, like, my question for you is, why is this so confusing, right? Because it does seem like, you know, the consensus vacillates from, yes, this is an AI bubble to no, it's not. When there are a bunch of developments, right? Like, think, think about, like, the turn of this year, the shift Went from, we're definitely an AI bubble, look at all the spending to like, AI can code. And so therefore there's. It's not a bubble because it's going to wipe out all of sass and more and capture all the value. And now we're like, oh, but like that sort of we that I don't. Don't want to say has plateaued, but that's been established and now there's going to be even more spending. So now it feels more like a bubble. So just talk a little bit about how when people think about whether there's an AI bubble, how they should think about it and why it's so confusing.

[B]

It's such a charged word. I think that's part of it. It's like everybody has a book and everyone's somewhat talking their book. And so it's like, you know, you have Michael Berry going out saying, like, it's a bubble. And he's a short guy. And so of course he's saying that he's like short a bunch of stuff. And then you have a lot of guys who are long and. And so they're saying it's all going to be. And so I think it's just hard, at least for me. This is what makes it so hard is like you're sort of trying to parse what everyone is saying, but also relative to what do they own and why are they saying what they're saying? And so I think it's really tricky. And then the word bubble sort of like has all of this emotional baggage associated with it. And so. And everyone's kind of nervous about a bubble. What is a bubble? And then I think you have this broader sort of Megatrend, which is like Silicon Valley has eaten the world in the last 50 years. So, like, betting against tech eating the world seems super, super risky and frankly seems like the wrong bet. And then also you look at AI and you say, man, if this is anything like the previous tech revolutions, wow, it's going to change the world. And then you try the technology. I remember, I think I was the first non employee to use Devon in 2024. The coding agent, right? If you were trying the technologies, you could see the future. You saw that these things were going to be extremely powerful. And so you try these technologies, you see that they're going to replace cognitive labor. And then you also look at some of these concrete use cases. You look at what Sierra is doing in customer service, you look at what Harvey's doing in law, you look at what juicebox is doing in recruiting, and you think, wow, like, these jobs are all going to be different. So, again, you're at the front lines, you kind of see that everything is going to change, and then you're trying to forecast that out, and there's a timeline to get there. And so I almost. I sort of avoid this kind of bubble conversation because the bubble conversation becomes very myopic. Suddenly everything becomes about one, two, three things, and people almost stop thinking. Like, you say the word bubble, and people just. Their brain shuts down and, like, maybe they're. And they're on one side or the other, basically based on where their financial incentive is for the most part. And so you just, like, can't have an intellectually interesting conversation. And the thing that I'm interested is, like, the nuanced conversation, because, remember, 80% of my job is I'm sitting in the boardroom of this company trying to figure out how do we navigate this AI cycle? How do we make sure that we're winners on the other side of it? That's what I'm spending my day doing is like, how do we win? How do we craft our strategy? I'm not a grandmaster like Elon Musk, but it's like, how do we work on our strategy so that we do a good job at the end of this cycle and sort of like, bubble. No, bubble is irrelevant. Like, at some point, there'll be corrections. We need to survive those corrections. In the long run, this technology is obviously going to work. We need to make sure that we're the winners, and we need to be massively aggressive to make sure that we're the winners on the other side of that. So I think, you know, it's better to actually have the nuanced conversation than to sort of shut things down with this kind of binary language.

[A]

And that's what we try to do here. All right, so let's. Let's keep going on the other side of this break. When we're back, we're going to talk about the greatest strategy game in the history of the world, as David puts it, and how the big tech companies are playing it. So we'll be back right after this. Hi, everyone. Alex Kanchwitz here. I want to tell you about a documentary I've made with Gravity to explore the future of AI agent security to find out if we're truly ready for autonomous agents. I sat down with MIT Professor Ramesh Raskar, former White House CIO Teresa Payton, Michelin's group chief, data, and AI Officer Ambika Rajagopal and Sharon Guy, a former executive at Alibaba. They each offer unique insights into this evolving landscape. We conclude with Rory Blundell, CEO of Gravity, to discuss the path forward, with Gravity leading the way. Join us on this journey. You can watch the full documentary at the link in the show. Notes. This episode is brought to you by DeepL when I sat down with DeepL's founder Jarek Kutliovsky on YouTube recently, we got into the case for specialized AI. DeepL voices what it looks like when the stakes are real time conversation. And honestly, it's something I wish I'd had for my own cross border interviews. Turning a language barrier into a non issue Deep Bell Voice delivers live translation in over 40 languages for virtual meetings and in person conversations, helping people speak in their preferred language without losing flow or nuance. Whether you're meeting with a customer, negotiating with a supplier, or collaborating with global colleagues, it keeps pace with you in real time, easily handling the technical terms, acronyms and product names specific to your business, so what you actually mean never gets lost in translation. And for the builders listening, Deep Bell's Voice API lets you embed real time speech, transcription and translation directly into your products. So go check it out for yourself. You can try DeepL voice for [e-mail] tryvoice that's DeepL.com tryvoice this episode is brought to you by AvePoint. Everyone's racing to roll out AI right now. Copilots, chatbots, agents doing real work. But here's the part nobody loves talking about. All that AI runs on your data, and most teams have no single way to see it, secure it, and prove it's under control. That's exactly what AvePoint does. For 25 years, they've been the trusted layer beneath the world's most demanding data. Now extended across your entire AI estate, your data, your cloud, and the agents acting on your behalf. It's how more than 28,000 organizations deploy AI with confidence so innovation scales without scaling risk. It's a single platform instead of a pile of tools bringing security, governance and resilience altogether. AvePoint the unifying trust layer for AI. Learn more at AVPT Co BigTechnology podcast that's AVPT Co Big Technology Podcast. And we're back here on Big Technology Podcast with David Kahn. He is a partner at Sequoia Capital. You could rate his writing at D Khan D c a h n.substack.com highly recommend you sign up. When David sends something out, I always make sure to read it. So let's Talk about this great strategy game, which is what's going on between all these big tech companies against each other. What if I were to like read to you sort of my one liner on some of these players and you tell me if you think this strategy is. This is the strategy and whether it's going to be a good one?

[B]

Yeah, I think this was my favorite post of 2026. And so I think in some ways the thing that's interesting to me is chess is an obvious strategy game metaphor. Right? So it's like, obviously people are going to think about this in terms of chess. I think the more interesting metaphors are like starcraft and Azad. And I think this will inform how we think about the individual company strategies, which is on StarCraft. For those who've played StarCraft, if not highly recommend at least reading about it. But it's like starcraft is a resource allocation game and AI is a resource allocation game. And so I think as we get into it, what are your resources and how do you allocate those resources? That's the most important question facing every big tech company. And then I think there's a question of there's three races in starcraft. What are your intrinsic abilities? We'll get into Google is maybe one race, Microsoft's other race, but what are your intrinsic abilities and intrinsic capabilities? And then the last metaphor I used in this post was Azad, which is this fictional game. In Ian Banks, the player of games, he imagines a civilization where you literally play a strategy game and whoever wins becomes emperor. And I think that's what's happening right now in AI. I think that the eight players around this 4D chess board think they are playing for the highest stakes in the universe. They are playing to be the emperor of the universe and whatever that means, the person who controls AGI, whatever that means. And so anyways, we'll get into the company specifics, but I think everything is downstream of what do the players believe, what are the resources available to them and what are the strengths that their sort of race gives them to act in this chessboard that is the world.

[A]

Okay, all right, let's go one by one then and see how far we get. For anthropic, I wrote the strategy is dominate enterprise, dominate code via products and API. So how would you rate that assessment and what do you think their resource allocation play is?

[B]

Disagree. I don't think that's the strategy. I think the strategy is corner the world's talent in AI and win, period.

[A]

Elaborate on that one.

[B]

I think it's been very clear Anthropic has had a very consistent strategy, which is they stand for something, they know what they stand for, and that enables them to recruit exceptional talent. And that talent compounds because in their view, and I think in a lot of people's views, Zuck clearly believes the same thing. Talent is the scarce resource right now. And so corner this scarce resource called talent, you got these people fighting for you and that's how you win.

[A]

Yeah. And it is interesting, like, if that is Anthropic's play, they actually lost far fewer people than competitors did to Meta during the Meta Superintelligence lab poaching spree back in the day. One interesting thing about that is let's say you go into a world where, let's say LLMs commoditize, right? Then you need to be able to win based on product. And if you have the best talent, then you could potentially have the best models and the best products. And that does seem to be down like sort of one assessment of where Anthropic is playing right now does that track.

[B]

But again, I'm going to push back on this. Like, okay, exactly where I think everyone gets confused because you sort of mix models here and you have to focus on what's the strategy. The strategy, and you can back test this back to before Anthropic was famously successful is like the strategy has always been extremely consistent. They deeply, philosophically believe in AGI. It is all about getting to AGI. They believe we are going to get there and they need talent to get there. And I think what the evidence at least so far shows that they have been able to push forward their frontier. And in their view, they're going to keep pushing for their frontier. And sure, they're going to have these products and it's great that they have products and they need revenue in order to attract investment dollars. But at the end of the day, it is all about AGI. That is my read. Again, this is just an outsider looking in. I'm not in the arena, I don't know. But as an outsider looking in, it's like very consistent strategy and they're going to keep executing that strategy. And sure, they need to like deal with this noise around commoditization and they need to do. But to them it's just noise. It's just like, let's just execute the strategy. Andre Karpathi just joined. All this good stuff is happening. Like, let's go. Right? I think that's. If you walk in the building and I don't spend hours time there. But I think if you walk in the building, that's what you sit next to. People at lunch, I don't think they're talking about like some China model. I think they're just like, how do we get the best people? How do we push the frontier forward? How do we make sure we have enough compute to do that?

[A]

Okay, so. But I would then say that the flaw in their model is that, you know, that can be. There's a chance that that is commoditized. Like the chance that AGI is commoditized is the flaw in that strategy.

[B]

What do you think that's your prerogative to. I'm not a Grandmaster. I don't like to like. To me, I think it's. I don't like to go say like, oh, that grandmaster is bad. I'm not a grandmaster. I'm just like a game.

[A]

It doesn't mean that they're bad. It just means, like, you know, look in the. Let's go. With the game of chess, when you move a piece one direction, it means that the other directions are shut off for you. So I think that if in this conversation, talking about, well, you move, it's a choice. You moved in that direction, you didn't move in that direction. That leaves your flank open. I think that if you want to talk about the game, it doesn't mean that you're a bad player. It just shows, okay, you've made that choice and that is your weakness.

[B]

But I think that this comes back to this question of, like, what are the fundamental resources available to you? You think that is where the chess metaphor breaks down and where Starcraft is more interesting? Okay, what are the fundamental resources that you actually can allocate? And at the end of the day, like, if models commoditize, that's good, obviously not going to be good for a startup that is driving all the frontier work. And so I think you have to bet on a certain world. Maybe that bet is wrong. That's why startups are risky. Maybe that bet is wrong. But, like, you do need to be opinionated. And I think what's beautiful about, you know, as a student of the game, what's beautiful about the game that Anthropic is playing, is that they're so consistent on that strategy. And if the assumptions they're making prove to be correct, they will be very, very valuable. Now everyone has to make assumptions, and generally those assumptions need to be based on the resources that you have available. And so of course, Microsoft is going to Bet on commoditization, because based on Microsoft's strategic position, that is a good bet worth making. And the universe is probabilistic. We don't know which. Which scenarios are going to play out. And so I, you know, the question I would put back on you is like, if you were Dario, what would you do differently? And the answer for me is I would do exactly what Dario is doing. I don't. I wouldn't do anything differently because I don't think you can play. You can't try to win in all probability scenarios. You need to try to drive the probability scenario that you believe is most likely. And this is where, like, I think probability breaks down. And I'm more of a chess player than a poker player, where I'm like, hey, let's drive to the end game that we want, right? And I think when you're on the inside of companies, that's how you think and behave. Wall street thinks of everything as a poker game. It's probabilistic. CEOs think in terms of chess. Like, I need to checkmate the opponent, I need to get to AGI. What's the win condition? How do I get there? And so I just think it's not productive as a chess player to be like, wondering, hey, is this, like, other thing going to happen? It's like, no. We just need to drive toward the outcome we believe in and we might be wrong. And that's okay. Business is complicated. The world is complicated. You're not going to be right 100% of the time. But if you get distracted, and I think this is what happens at big companies, so we can get into some of the hyperscalers where you have a committee around the table and everyone has a different opinion, that's actually much more challenging because you actually can't consistently execute a strategy, at least for Dario and I think for Sam and I think generally for startups and for Elon, for what it's worth, Dario, Sam and Elon are all executing strategies very consistently, and it seems like they are doing a good job.

[A]

Okay, this is funny, because what I wrote down for OpenAI actually sounds a lot like what you said for Anthropic. My line for OpenAI is build the smartest possible AI and figure it out.

[B]

Disagree again. I think OpenAI's strategy is everybody underestimates AGI, all in AGI, and we're going to be the most aggressive and everybody else is going to take fewer risks than we do, and we're ultimately going to be proven right. And Win. And I think if you look at their compute strategy, that seems like what the strategy is. And again, that's a very coherent strategy. Sam talks about this all the time. These things are exponential. People underestimate exponentials and we're going to get there. And by the way, I think OpenAI has the added advantage, back to my resource thing of what are your resources? Where OpenAI invented the field. So I think there's always an invented the field advantage. I think they have incredible talent in there. If anything, Zuck's attempt to compete with OpenAI and just the failure of that so far just demonstrates how good the talent slot opening is. And I think Sam's strategy is very coherent, which is like, we're just going to be the most aggressive.

[A]

Okay. So I think is what you're saying about OpenAI kind of like they're. Yes. Anding anthropic. Like they're like, yes, we're going to drive towards AGI and we're going to try to build more aggressively than you.

[B]

Yeah. And I think that's the plan. That's, that's who Sam is as a leader. I think that's been very, I mean, again, I always try to rewind the clock because everyone like forgets. But like, remember that he raised $1 billion for this lab when it was like, that was a crazy thing to do. Right. And then he raised $10 billion for Microsoft. He's always been, I think you have to back test these strategies and ask yourself, when I look at the prior decisions, does this confirm or does this violate my assumption about the strategic thinking that is going on in these people's heads? And these are humans, right? They're complicated humans. They're humans whose impact is going to have global scale. And so I think it's fair that we analyze their decisions, but they're humans. And I think if you back test Sam's previous decisions, he's been very consistently a big believer and he's very consistently been the most aggressive player around the board.

[A]

Yeah. All right, so if you don't want to talk about the weaknesses, I'll talk about them. And I would say if you're making that move right the flank that you leave open is what we were talking about earlier that duration. Mismatch, where if you're going all in harder and faster than anybody else, you're the most susceptible to potentially having that timeframe not line up with sort of the, the, the, the technology's time frame in your business time frame. Have a mismatch.

[B]

Okay.

[A]

I'll say that and let's move on to the next, the next company, which is Google. So. So my one liner for Google is this is really fun, by the way. So thanks for playing it, playing the game. This is a game within a game,

[B]

like commentators watching the guys on the field, right?

[A]

It really, it does feel like that. Okay, so for Google, mine was. Mine is change search just enough to keep our customers and cross our fingers. Also try to build good models and cloud services without killing each other.

[B]

Google is hard and, well, so we'll spend some time and Google is hard because it doesn't seem to be run by a single individual with a single vision. And I think there's varying degrees to which this is true for the hyperscalers. But let me tell you what I would have the Google strategy be and then let's try to understand what it actually is. Google's number one. Google has two big advantages, right? It has a cache machine from search that can fund a lot of things. And then number two, it has tpu. And I cannot overstate what an advantage it is to have tpu. They've been building this chip for a long time. It's a really good chip. Some people think that part of why Anthropic is doing so well is because they're using TPUs. And so if I was Google, I would be all in tpu. And I think they are to an extent. Right. But they're right now hoarding the TPUs for themselves for the most part. They're starting to open up the gates to let other people use TPUs. I think something I've been surprised by is you look at Jensen and I think one thing you got to appreciate by Jensen is the man has been doing this for 30 years. And for 30 years he's been really consistent on one thing. Ecosystem. You go to Taiwan, Jensen's a national hero. You know, I can. There's probably 100 people who. Jensen's just like made rich because he doesn't care about the incremental point. He's like, we're all going to win. It's all going to go well. Jensen makes other people rich. I think that's one of the best things about Jensen. He's not, I'm going to be the most valuable company in the world. You can have some, some, some juice too, okay? And so Jensen sort of builds this ecosystem. You look at Cuda, you look at sort of the ecosystem he's built around the chip. It's like if I was going to go into business tomorrow and start A company I'd love to work with Jensen. Jensen is going to treat me well, he's going to be a good business partner, et cetera, et cetera. Google has this sort of like more insular thing, right? Google for the last 30 years have built everything themselves, right? Like everything is built in house, everything is invented here. Even though they invented all the core technology behind AI, they couldn't really figure out how to do anything with it. Now they've been good at acquisitions historically, so they've acquired a lot of companies and they're good at acquiring talent and their technical team is really good. But if I was running Google, to me it's like the all in TPU strategy just seems so smart. It's like, hey, Nvidia is a $5 trillion company. We can build a multi trillion dollar company just on tpu. But Google. And now let's come back to what is Google actually doing? Because they're hoarding the TPUs. I think they're actually doing the Go Chase AGI strategy and I just don't know if they're well set up to do that because I think they are actually going down the OpenAI strategy more and more, right? It's like we're just going to keep the TPUs for ourselves. We're going to try to get the best researchers, we're going to try to make the best breakthroughs. Gemini is going to be amazing. Gemini is going to compete and maybe that's the right strategy because maybe AGI is such a big prize that it's just not even worth some kind of hedge bet or some ecosystem bet. I just struggle sometimes where Google is one where I said like Darto's playing a beautiful game, Sam's playing a beautiful game. Like as a student of the game, I can appreciate beauty in the game and to me it's beauty, right? It's just like, wow, what an amazing player that is. Google, it's an amazing company, it's got amazing resources, it's got amazing advantages. So in some ways Google is the easiest to appreciate as a company. But then when you think about the gameplay, it does sometimes feel disjointed. And so I don't know anyways, I don't have as clear of a view. I was very opinionated on the other two. I don't have a clear as a view at Google on what they're doing. They're spending, you know, $200 billion in capex, they're accelerating their, their, their cloud business. I think one thing to point out, and I Think sometimes there's like some finance one to one that kind of gets lost in these conversations where it's like, you know, if I build a house and then rent the house to you, that's capex with revenue. So of course I can get rent revenue by building something. But you have to ask the question of like how much revenue are you getting? And then for these hyperscalers, it's not just that they're building the house and then collecting rent, the revenue goes up, but they're actually building the house and then investing in you so you can pay them rent. And so I think there is this broader question on the hyperscalers which is like the resource they have is cash. They're all using it to the maximum degree. And is it almost this curse of like, I almost wonder if there's like a resource curse for these big hyperscalers where they have so much cash so they're incentivized to spend it when really the thing that's going to drive success is like not cash. And cash just can only get you so far. And sometimes I look at this capex boom and it's like, I think it's almost downstream of this resource curse. It's like, oh, I have so many resources, I have to spend those resources. I'm in this game theory with Amazon, Microsoft and Google all have these cloud businesses and it's an oligopoly and it's like cash cow, right? It's like the best business in history probably it's the greatest oligopoly in history. It's producing all this cash. So I kind of have to compete with the other guys anyways. You can kind of like pull the thread and you sort of get to a place where it's like this strategy seems downstream of my resources available to me as opposed to being downstream of some outcome that I'm trying to get to.

[A]

What is the alternative though for them? Like is it to sit it out?

[B]

I think that's where I said like what I would do, you know, and maybe this is wrong and I'm not in the building. I don't understand all the constraints that you have. But to me it's like I would try to build an Nvidia competitor. I just think the TPU is a wonderful product. If I could have 50% market share in 20 years of AI chips, that's a pretty good business. I would sort of have a management team that does that now. Sure, my internal team can be a customer of that business. And there's advantages to vertical Integration. But one thing, and I've been writing about vertical integration for a couple years, the reality is when you look at the ecosystem today versus two years ago, the vertically integrated companies have not been able to leverage that vertical integration into some model advantage. And so at some point, and this is where I think a great chess player has to change strategies or a great StarCraft player. Like at some point you get data, like I think there is data that vertical integration isn't resulting in the advantages that these companies thought it was going to drive. And so, okay, like maybe, maybe there's a different strategy. And I think there is a coherent strategy that's like, we're going to have one team, Chase AGI and we're going to also monetize this TPU asset. And then there's the search question, right? Like in some ways if you do the old school, you know, there's Alastair Nairn, who I love and think very highly of and who wrote the book the Engines that Move Markets, which is like the great book on technology investing, who says it's easier to short the canal than it is to be like, figure out which railroad's going to win. It's like, is Google search like the canals, right? Is Google search just like, you know, these ad driven businesses, part of why I think Meta and Google are so all in AI is their businesses are the most at risk from AI. Like if we all move to chatbots, if we all move, if our time eyeballs move to these other interfaces, then these ad driven businesses have a lot of questions. And so when you talk about what flanks you have, I think ad driven businesses are a very risky place to be. So anyways, I think it's somewhat coherent what they're doing. I just think all of these businesses, these big hyperscalers, there's a lot more questions about and part of it is like, and I think this is the beauty of founder run businesses. Easier for a founder. Like look at Zuck, his strategy is coherent, we can get into it. He's like all in, right? There's like one strategy which is like spend the most money, acquire the best talent, who cares? Founder led businesses can just act and behave differently than non founder led businesses.

[A]

Yes. Okay, on the one question, on the vertically integrated businesses not being able to produce the foundational like the leading models, why do you think that is? Like, so basically what you're saying is the companies who like their main business isn't necessarily selling the AI, it might be, you can put it into play somewhere else. Like if you're meta. You can put it into play in a consumer product if you're Google, maybe in Google Maps or Gmail, right. But they're all trying to build their own models. They've struggled to, to sort of compete with those that are not vertically integrated. So basically have to make the money on the model itself. What do you think has been the source of that struggle?

[B]

Yeah, it's funny, I think like every 10 years, you know, the Harvard MBA professors change their mind and like, is vertical integration good or bad? And so, you know, maybe just first principles in it, right? Like there's elements in which vertical iteration is good. You look at Tesla, you look at hardware companies, vertical iteration typically is pretty good in the hardware supply chain because you have all these supply chain partners and your supply chain partners maybe don't have, you know, you look at SpaceX, right? Like you had to vertically integrate. Like your supply chain partners were bad, everything was too expensive, the math never worked. And Elon just squeezes. Vertical integration. Vertical integration. Vertical integration. And so now vertical integration is super hot. Because Elon has done such a great job of vertical integration. Let's propose the hypothesis that vertical integration between software and hardware, which is what they're trying to do with the chip and the data center and the model, isn't necessarily a good thing. Like it's just a real estate thing. I can rent the real estate thing, I don't have to buy the real estate thing. And now again, I think you could debate this. I don't think this is clear. So this hypothesis, right, There was an argument, and I thought this was a good argument when it was first being made. Just haven't seen the evidence. There was an argument that if I control the data center, then I can make a better model. That was the argument. When you look at the evidence over the last two years, it doesn't seem to be the case. And maybe it's because these companies are so big that the guy building the model is like over here and the guy building the data center is like over there. And they never talk to each other. So you might as well. They might as well be different companies. That would be my hypothesis is that like in reality they might as well be different companies. And so if they might as well be different companies, then aren't you advantaged just buying whatever the best chip is? Aren't you advantaged just like using whatever. The free market exists for a reason? The anti vertical integration argument is capitalism, right? Like capitalism emerged where you have all these specialization and like, you know, most of the trend lines of capitalism is you specialize in the thing that you're really good at. You have a relative advantage at. You don't have a relative advantage. You let some other guy who has a relative advantage win that business. And, you know, maybe Apple is sort of the anti vertical integration, right? Apple assembles the iPhone, somebody makes this chip, somebody makes the camera, somebody makes this, somebody makes that. And I just make sure the product is amazing. And Apple is better off buying the camera from some guy who makes the camera, because there's 30 guys trying to make the camera for Apple and I get to pick the best one. So that seems to be the evidence so far, but we'll see. I think the evidence is not in yet in a way to be definitive on this.

[A]

Okay, all right, let's do a few more. So for Meta, I have commoditize our compliments, spin the wheels until someone builds a good consumer AI application and then copy it and distribute it.

[B]

I'm a simplifier, so maybe to a fault. So I'm just going to give you. All of mine are very like, good, good, right? Which is just like, buy talent, it's a mercenary army. Buy talent, you can pay enough and get the people that you need. And again, the question. So the question for Meta is like, can a mercenary army do as well as a missionary army? Obviously, a mercenary army. We're buying these companies insane golden handcuffs, billions of dollars for AI researchers. I happen to think that the people they've acquired are very good and that the team is very good inside of Meta. That's my personal opinion. And so I happen to think that there's a good chance that they do figure it out and that Zuck's approach is coherent. Again, it's like, what resource do you have? Well, you have Instagram. It spits off cash. What can you do with that cash? You can buy Alex Wang. And then the question is like, okay, well, does that get you far enough? And then the emergent question in 2026 is, do you have this cultural dysfunction where you have this organization with tens and tens of thousands of people and then you have these 50 people off in the ivory tower working on the thing that matters? Clearly. And you see all these leaked press releases and whatnot. Clearly there's this organizational dysfunction that has emerged as a result of this. And so I think the question for Meta is, what happens? And then the other question, you know, when I graduated college, Meta was like the hottest place in the world to work, and today it's not. And so the Other question is like, do you, you know, are you fighting a losing battle against organic talent flows? Is like, do you need to win the organic talent flow game? And it is a question for all the big tech companies, which is like, organically, the talent flows are not there. I spend a lot of time, we can talk about talent. I meet 300 young people a year. I spent a lot of time on talent. I'm a very talent centric investor. The organic talent flows simply are not there for the big tech companies in a way that 10 years ago really good people did go to work at these companies and today they are not perceived as leading edge companies.

[A]

For Meta, if they figure it out, like let's say everything goes their way, doesn't it not really matter if there's no like consumer widespread, like consumer application of AI, like super personal, super intelligence, or do you feel that if they're able to succeed in their strategy of this, you know, buy talent, hoard talent, then naturally that personal superintelligence will emerge.

[B]

This comes back to like, I think we're all underweight AGI and I think these grandmasters are playing for AGI. Zonk is obviously playing for AGI. And so whatever that means. I think it's, you know, I don't know if we have time to go in and like try to define AGI. There's no one good definition of it. But whatever it is, you know, this, this definition that my partner Constantine proposed, which was basically you go from 1% of cognitive labor done by machine to 99% of cognitive labor done by machine. I think it's probably the easiest definition because that's what happened in the industrial revolution. 99% of labor was done by humans. That 99% is done by machine. Whatever version of that future is, you know, Zuck is going to be a very wealthy man if Meta wins that race. And shareholders will be rewarded if they win that race. And so I think that is, that is the path they are on. And again, I think it's coherent. Whether it will work is another question. Yep.

[A]

Okay, so here's mine for Microsoft. I basically have it that Microsoft is going to try to knock OpenAI and anthropic down a peg, which I think we've kind of talked about, is turn those models into a commodity and leverage existing enterprise relationships for profit.

[B]

I've always had a lot of respect for Satya Nadella and I've always sort of, I think, had this pretty optimistic view on Microsoft. I think they don't get enough credit for the fact that they own so much of OpenAI. Right. Like, the deal making that Satya has demonstrated in his career as Microsoft CTO is just unmatched. Yeah.

[A]

It's like 27%.

[B]

Yeah.

[A]

Of OpenAI.

[B]

In the beginning, I think it was more. Maybe that's what it is now. But in the beginning it was a lot. It was, you know, there's these stories of, like, Sam went to all the big tech companies, I'm pretty sure, and offered them this deal and Satya was the one who did it. And so I think there's this question of, say he doesn't get enough credit for the fact that he's played a very, very good chess game. And again, to my beautiful players, comment, like, I think Saya is one of these players that you just have to stand there and go, wow, this guy is thinking at a level that we're not thinking. And I think, to your point, to some degree, I think Microsoft's strategy right now is like, whoever wins will win, you know, and I love that. One of my favorite things about business, I think the best CEOs that I want to invest in are people who, no matter what happens in the world, I'm going to win. Alex Wang is one of the guys. No matter what happens, he's going to win. And it's just because you're nimble and you move fast and you're adaptive and you're aggressive. So I think Syed has had this strategy. And when I look back at 2025, when he attempted to kind of pull back on CapEx and had Oracle do a lot of the CapEx, I thought that was a pretty smart strategic calculation. He also updated pretty quickly that this thing is taking longer than he expected. And so he still has to be in the game. And so when I think about a company that's playing its cards just really well, I think Microsoft's distribution machine is amazing. So Microsoft benefits if OpenAI and the labs win, and it's all frontier models and it's not commoditized. He owns a lot of that. He's going to do great if they get to AGI. Great. Owning 20% of AGI is pretty good. And then if it doesn't and it commoditizes, he's going to win the distribution game. And so I don't know. Saya, I think, is one of these players who. He's sort of slower. He's not, you know, you don't see Saya publishing a tweet every day about some new thing they're doing. He's like a little bit more slow, you know. Again, on my comment at Google, I said what I would do. I think I would probably be more aggressive if I was him on M and A. Like, he's been a great buyer in the past. He's, he's, he's someone that I think knows how to get the most out of companies. So I've been surprised they've been less aggressive on M and A. But again, like Saya is a better chess player than I am. He, I think he does have a clear master plan here and I think he's been extremely prudent and aggressive in how he's operated in a way that I really admire.

[A]

So here's mine. For Amazon, it's do very little sell compute profit.

[B]

That's probably the closest one that I agree on, you know.

[A]

All right, good. Look at us.

[B]

We're almost there. What is Amazon strategy? It's almost. Again, it comes back to this like strategy by committee thing where it's like hard to know exactly what the strategy is. Certainly the output of that is like, do very little. They have their internal kind of AGI lab. It doesn't seem like it's a big priority for them. They have a huge cloud business. Their cloud business does seem to be losing share to Azure and Google over time. I mean, it was the best of the cloud businesses in the software era. In the AI era. Will it be quite as differentiated? I'm not sure. So I think Azure and Google and GCP have kind of caught up and have some pretty good advantages on AI cloud. And so you look at the cloud business and you think, I think your analysis is probably correct. It's like, hey, we're just going to along for the ride. We're going to do kind of what everyone else is doing. We're not going to do anything crazy. We're just going to do what everyone else is doing. We're the best at building data centers. Maybe to give them some credit on the things they're really good at, like, they're the best in the world at building data centers. They have the best cloud business in the world. They know how to build a cloud business. And kind of steady as she goes on all these earnings calls, they basically talk about like, hey, we're really good at building data centers and that's an advantage. Again, I'm waiting to see those advantages kind of materialize. I think it's a good argument, but let's see it materialize again. Maybe we're going to fast forward the clock in 10 years. And it's kind of the simple strategies did end up working well. Obviously there's less intellectually interesting about Microsoft strategy. It's just like, and the thing is this do nothing approach is not really doing nothing. They're blowing hundreds of billions of dollars on capex. So that's the other thing that's kind of weird about the hyperscalers is the baseline is so high that you can't actually say you're doing nothing. Like you're spending hundreds of billions of dollars, you're spending all of your investors free cash flow. And I think this is something that does happen in these kind of crazy cycles is that our baseline expectation for what's normal adjusts. So it's like, oh yeah, spending hundreds of billions of dollars capex, that's like a do nothing strategy. That's actually insane. A lot of money that you're spending. And so in some ways I worry for Amazon which is like heads I lose, tails I lose. It's like if this all goes well and AGI happens, they're not well positioned for that. And then if it doesn't happen, they're kind of in the bath with everybody else on the capex that they've spent. And so it's like where's the edginess, where's the spike that makes you think like man, this is a great strategy. So again, I don't know, there's some strategies I understand more than others. I assume these companies have better strategies. By doing, the founder led companies have some advantage in their ability to be aggressive and opinionated. Whereas you have this kind of committee washing in these big companies. Apple is the one by the way, with the do nothing strategy. Like you got to look at Apple and say wow, like there's some chance that Apple just comes out looking like a genius on this. If it does commoditize. Apple's not spending money on capex. Apple really is the do nothing strategy. Obviously investors are penalizing that for them today. But you can imagine the scenario where Apple looks great in 10, 20 years because of the decisions they made. And so Apple in some ways is also an opinionated strategy in a way that some of the hyperscalers are not. And I think the reason the hyperscalers can't be opinionated, coming back to this game theory thing is they have this golden goose where they fight with each other and it's this amazing oligopoly. And I do think there's this winner's curse or resource curse in competitions where you're actually saddled by the fact you have this cash machine and you just have to use it because it's the thing you have. Whereas everyone always said about anthropic, like they're on a knife's edge, you know, if they lose their frontier, that's bad for them. But people perform at their best when they're on the knife's edge. And I think that has been the empirical evidence of the last two years.

[A]

Nvidia. This is probably too glib, but I just wrote pray that inference isn't commoditized.

[B]

I would say Nvidia prop up the AI ecosystem and this is very consistent with Jensen's history and I think to his credit, like a pretty smart strategy. Jensen just needs AI to work. If AI works and doesn't crash, Nvidia is going to be in a great spot. It's one of the most valuable companies in the world. Jensen is one thing I like to talk about in AI in 2026 is like the AI trade has sort of degraded in the last three years. Like 2023, it was the year of Nvidia, right? Nvidia is like the greatest company ever. It's an amazing chip, it's an amazing technology. Cuda's amazing. He spent 30 years building this thing. It's like when you think about 2023, it was like this amazing crowning moment for Nvidia. 2024 you get Broadcom. Hock Tan is one of the all time great CEOs in business history. Broadcom's a premier brand. It's the number one design firm in semiconductors. Of course Broadcom is going to do really well. 2025, you get GE, Vernova, you get Vertiv, you get Siemens. You basically have these industrial giants that step in and look, they're good businesses. GE is an American national hero, Siemens is a German hero. They're not quite as good as Broadcom or Nvidia from just like a raw amazingness perspective, if you will. But fine. And then 2026 you have like SK, Hynix, Micron and Samsung which are like right place, right time, price hikes just like squeeze margin out of the hyperscalers. So he's on Nvidia. Like I just think it's worth saying it's one of the greatest businesses in human history. And so he just needs the thing to keep going. And I think that circular strategy where he's investing in everybody and trying to make sure everyone else wins to some degree. I think it's like Jensen is leaving so much money on the table. And it's because he genuinely, at the bottom of his heart, I think, just wants this whole thing to work and I think he probably should get a lot of credit for that and obviously doesn't. You know, Nvidia stocks barely move this year even though these high flying beta companies are going crazy because they're raising price. Jensen isn't being insane on price. He probably could have been. He hasn't been crazy on price. He's good to the ecosystem. And so I think anyways, ecosystem first is my summary of what Jensen is all about. And I think he's been about that for 30 years.

[A]

So this idea that he'll like give startups GPUs in exchange for like percentage of their company, that's basically his. If you have a chance of working in AI, I'm just going to make sure that you at least get a chance to take that swing and I'm going to give you the most precious commodity and that's kind of part of that strategy.

[B]

I think sometimes people's egos get in the way of doing what's right for them. And Jensen is someone who I think is the opposite. He's pretty low ego about it all. He's like, hey, I'm just going to make everybody win. Maybe I'll leave some dollars on the table. It's all great. We're all going to do great. AI is going to be amazing. It's going to change the world. Every consumer in the world is going to have a better life because of AI. I do actually view Jensen as kind of the make the pie big in a way that everybody else is kind of fighting over share and this and that. I think Jensen just wants the pie to be really big. For what it's worth, I think that's why he's so obsessed with open source. He just wants the pie to be big. He just wants AI to make people's lives better. I think he's very well intentioned and I think it's because he's think about how many years they spent building Cuda for this moment. Like in some ways, Jensen made a bet on AI. A long, long time ago. Jensen was one of the first people to make a bet on AI and I think that he can. You can't change who you are. That's maybe why. Also whenever we talk about these companies, I come back to the human personalities and the game theory. What's their world model? I have my world model which I've shared on this podcast. What's their world model? I think you can't change. At some age, you're not changing your fundamental world model. And Jensen's fundamental world model is you win, I win.

[A]

Right. Okay, so for SpaceX, is SpaceX kind of pursuing the strategy that you wanted for Google? Not they don't have their chip, but they started to try to build their own model. And they're still building it with Grok, but I think they realized that they weren't, you know, they're sort of all in pursuit of AGI through their own model, might not have been the winning horse, and then pivoted to being a cloud, an AI cloud or a Neo cloud. And that's sort of how they're going to live or die here.

[B]

To me, SpaceX is like a meta commentary on financial markets, which I love, which is like, you know, I mentioned earlier, like, Silicon Valley is eating the world. Over the last 60 years, like, it's been amazing. We have chips, we have Moore's Law, like so many good things have happened and then we like, have resulted in like, you know, Instagram and TikTok and whatever. You know, it's just like, not. I don't think that the hope for humanity element has been a little missing. And Elon is almost this like, meta commentary, which is like, if you move civilization forward, then you like, make a lot of money. But he really cares about moving civilization forward again to the underlying incentives. I don't think Elon is fundamentally dollar motivated. I think Elon is fundamentally mission motivated. And so let's go to space, let's colonize other planets, let's advance human civilization. AI is a way to get there. We're going to have robots. Elon, in some ways is working on every important idea that matters for civilization. It's like we need tunnels, so we're going to have the boring company and we need neural link, so we're going to have neuralink. And. And he's able to aggregate talent. And I just think his leadership style is so masterful. Right. He attracts the best people in the world. He makes them want to do their best work and he cuts all the blockers. You know, I mentioned management by committee. It's like great performers don't want to be managed by a committee. Great performers want all the blockers to go away. Let's just focus on achieving the mission. And you meet the veterans who are at Tesla for 15 years. They're so proud of what they achieved. My partner, Ravi Gupta has this thing, he says, like, the meaning of life is earned achievement. And isn't Elon all about helping other people achieve that meaning of life. What an achievement you feel when you were part of one of these companies. And so Elon is just so consistent. To me, Elon's about chasing what's good for humanity while bringing out the best in the human beings who work in his organizations. And obviously they're intense and obviously people come and go and it's imperfect. But SpaceX is pursuing the biggest missions for humanity right now and it has the most optimistic vision of what I do think Elon is correct in a lot of his commentary on like, he is pro human. He's like fundamentally the most optimistic pro human, pro progress vision out there. And that's why I think retail has lined up behind him.

[A]

Okay, I want to end here and this is sort of a turn from where we've been the entire conversation, but I'm so glad that you wrote about this because I don't think it's talked about enough. And this is something that you wrote about how AI might change spirituality. And you even asked this question in the extreme. Will humans worship AI and what would such worship even look like? And you know, I mean, obviously the history of spirituality is humans worshiping what they think is this higher power. So it would go to stand to reason that like, if we invent AGI or super intelligence, there will certainly be some people who will be like, this is the manifestation of the like, effectively, like we've created the creator. So I. Which is crazy and weird to think about. So I'd love to hear your, your thoughts on, on where this goes. And also just as a corollary to that, how the pursuit of building a artificial brain may change the way that people think about religion itself. At least the traditional forms of religion that we have on this planet today.

[B]

Great question. It's a deep thread to pull, so I'll try to give an answer to it in a longer conversation. Obviously sometimes I sort of have these side quests that I go down and it can be multiple year side quests. So I'm just going to say that out loud. These side quests don't always converge on the main quest. The main quest is technology investing. How do we build the future of humanity? I'm motivated by intellectual interest. The reason I got into AI almost 10 years ago is intellectual interest in what AI was and what it was doing and the potential. I sort of had this feeling. It's funny to think back to those times at this feeling of like, man, it's like so crazy that we have all this data and we just make dashboards with it. You know, like that's what investing was 10 years ago. And then and I had this view of like, maybe we could do more with the data, right? It was really that basic of an insight. And I pulled that thread and the nice thing is like you pull these threads and then some of the threads are longer than others. And look, this AI thread has taken us 10 years and we're still pulling the thread and we're going to be pulling the thread for a long time. This other thread, and this is a personal interest, not a career interest, is like spirituality and God and religion. And I grew up religious and believe in God. And so what is God, right? And I think when I invest, you sort of try to invest in like emotionally mature humans. And so part of our quest I think as individuals is like to become emotionally mature ourselves. And part of becoming emotionally mature I think sort of trying to understand the universe. I think this is why a lot of physicists go, you're trying to understand the universe. And religion in my mind is this like, you know, thousands of years quest to understand the universe. And I have a belief that, you know, 5,000 years of wisdom probably got us somewhere and that we were making progress, that humans, I don't think our IQs have like fundamentally changed that much. Humans were pretty smart. And what humans did is they constructed whatever we call religion and they constructed these systems to try to get at ground truth in the universe. And in my view there's like all, it's all pointing at one truth. I think a lot of people would believe this. Like you can read the Quran, you can read the Torah, you can read the Gospels, you can read the Mahabharata. Like there's kind of one truth in the universe. A lot of people have been trying to get out what that truth is. We call that truth God. Anyways, this is a side quest, right? There's nothing to do with AI and yet it really informs our day to day lives, especially in the valley. Like the valley is this place that is so devoid of God and you can pull the thread of like, why is America not just the Valley? The Valley is an extreme version of this. Why is America so empty of godliness, right? And then you kind of pull the thread and you go back to Freud and Freud basically said like, hey, actually God died. Freud didn't kill God, he just noticed that God was dying, that religion was dying in the West. And Freud invents therapy, you know, therapy and therapy culture. If you look at the last 20, 30 years, therapy culture has like, sort of taken over. And yet you look at the stats, people are lonely, people are depressed, people are unhappy. And so I actually don't think these are parallel threads. Like, AI is happening and then there's this broader cultural thing that's happening. And frankly, people don't really want to hear from me on the cultural thing, which I get. It's like, I'm just a VC who's investing in startups. Fine. But I think that the cultural thing has echoes inside of this AI universe, where I think one view of it is it's the Tower of Babel quest. Right. We're trying to build God. That didn't end well in the Tower of Babel. What happens? How does that play out? So I think that's one view, and then the other view is that, hey, God died in the early 1900s, just as societal level. And then these people are trying to create a new God and maybe they will successfully create a new God and maybe that will reinfuse the world with some spirituality, And I think there's going to be a lot of debate along the way of is this good, is this bad? It's hard to have an opinion. I'm still in the, you know, I'm in the 10 years ago on AI. Like, I'm in the early phases of pulling this thread. But I will say, the more I've pulled this thread, the more I've learned. I think it's intellectually extremely deep area. There's a lot of people who've written about it, and I'm probably in inning one or two of learning more about what I would call transcendence. It's the transcendence quest, like spirituality and religion is one version of that. But I think we all sort of crave transcendence. And in some ways this AI quest is sort of fulfilling this need that people have for transcendence. And I wonder if that is part of what's powering it, why it's become so big. I wonder how that plays out as the technology evolves.

[A]

Yeah, no, it's crazy to think about. And I think you're right. Like, it isn't talked about in terms of AI development, but there's no avoiding it. I think it has to play some role when you're trying to create intelligent life on your own, whether you believe in God or not. There is a. Some spiritual parallel there. So there's this question that we all

[B]

share in common, which is like, how do we heal society? Like, society is hurting right now. How do we heal society?

[A]

Okay, do you think AI could do that?

[B]

I don't know. I think there's elements in which it's been good and there's elements which have been bad. Right. And I think we, you know, to some degree, I think that exacerbating loneliness is concerning. Certainly technology has made people very lonely. We don't have. Communities are breaking down. We don't have. People don't have a sense of community right now in this country. And so how do you do that? I think there's one argument which is like, return to the religions of old. Maybe that will work. I don't know. Right. It's. As a technology guy, it just always seems hard to go backwards. So how do you go forwards? But then you have all, you know, you have Peter Thiel saying that it's like the Antichrist. Right. So it's like, what I don't know is the. Is the answer. And this is why I'm pulling this thread. And I think it's interesting, and I hope that through developing this technology, we come out with some optimistic answer that enables people to have enriched lives. And I do think a lot of the people who are building this technology believe that once we're freed from daily labor, you know, we're going to be able to have these rich lives. I do think there's a Jewish phrase which is, there's no work without Torah and there's no Torah without work, meaning, like, we need the spirituality crest and we need to be grounded by some real things in life that we're working on. And so I think this dual quest is in some ways, like the dual quest of humanity. It's like you need to earn your daily bread, and then you also want some version of transcendence.

[A]

Well, I hope you keep writing about this, and folks, I do urge you to go sign up for David's newsletter. DCON D C A H N. Substack.com I'm looking forward to reading more of your writing, and I know you don't do this often, so really appreciate you coming on the show, and I hope we'll see you again. David, thanks for coming on.

[B]

Thanks, Alex. Thanks for having me. Fun conversation.

[A]

Really fun. Really fun. One of our better ones here. So thank you again and thank you all for listening and watching, and we'll see you next time on Big Technology Podcast.