Sequoia Partner, David Cahn on Who Wins in AI, Defence & The New $0–$100M Playbook

1. 0:00 – 1:09

   Intro

   1. DCDavid Cahn0:00

      I do think we're in an AI bubble. You can see the fragility. Everybody can see the fragility. The thing that I think is more interesting is, who's gonna survive the bubble? Consumers of compute benefit from a bubble, because if we overproduce compute, prices go down, your COGS goes down, and your gross margin goes up. The lesson that punches you in the stomach in venture is you can't make a company succeed.

   2. HSHarry Stebbings0:20

      How would you respond to Sequoia were asleep at the wheel when it came to defense, not being in Helsing and Anduril, the two clear market leaders in the category?

   3. DCDavid Cahn0:29

      I would say... (mouse clicks)

   4. HSHarry Stebbings0:32

      Ready to go?

   5. NANarrator0:32

      (instrumental music)

   6. HSHarry Stebbings0:44

      David, I love your writing. Our episode last year was one of the most downloaded shows. I had, like, the CMO of Meta tell me that it is the single show that he has forwarded to more people and cites more often than any other. Not to make you, you know, nervous or set the pressure for this episode.

   7. DCDavid Cahn1:03

      (laughs)

   8. HSHarry Stebbings1:04

      But thank you so much for joining me again, dude.

   9. DCDavid Cahn1:06

      Thanks for having me, Harry. You're always very kind.

2. 1:09 – 10:36

   Why Building Physical Data Centres is a Moat

   1. DCDavid Cahn1:09

   2. HSHarry Stebbings1:09

      Now, the Year of the Data Center sounds wonderful. We had an amazing discussion last year. What did you predict last year, David, that happened and we are seeing in action now?

   3. DCDavid Cahn1:21

      I think there's really... So we talked about, last year, this concept of steel servers and power. And I think if you remember, you know, rewind to summer 2024, the big conversation at that time was compute models and data. That's what everybody was talking about. And I sort of had this view that everyone was underestimating the physicality of these data centers. I'm on the front lines. I'm talking to people every day. And you, you know, you talk to people, they're flying electricians to Texas, and they're trying to buy out generator capacity, and, you know, generators are sold out until 2030, and, and so how do you get in line, and how do you do that? And so I sort of had this sense that people were thinking very abstractly, sort of in a, in a bits perspective about AI, but they should be thinking in an atoms perspective about AI. And I think that prediction came true in two ways. Uh, the first way is, the best trade of 2025 was the AI power trade. A lot of Wall Street people made a lot of money betting on the fact that power was gonna be the constraint and we were going to move away... You know, you hear Sam Altman now talking about gigawatts every day. He's not talking about dollars anymore, right? So we're moving away from dollars, and we're moving toward gigawatts, and I think that transition has s- fully happened in the last year. The second way I think it was right, and I saw... You know, it's funny now, like a year and a half later you see this on the cover of The Economist and the cover of The Wall Street Journal and the cover of The Atlantic. The mainstream media has now really picked up on this narrative of the physicality of AI is what translates to GDP. I mean, GDP is an imperfect metric, and it generally captures physical things more than virtual things. And so GDP now is picking up all of this construction boom that's happening, all this deal that's getting created, all of the physi- physical stuff that's happening in, in the AI data centers, and you're seeing these stories which I think are true, which is, AI is now one of the biggest contributors to GDP growth in the United States, and so I think that's the second way in which that prediction has played out.

   4. HSHarry Stebbings3:03

      Does its contribution... You know me, I just go rogue and go off grid, but it's much more fun.

   5. DCDavid Cahn3:07

      (laughs) Much more fun.

   6. HSHarry Stebbings3:07

      Does its contribution to GDP growth go contra your $600 billion question in terms of where the revenue will come from?

   7. DCDavid Cahn3:16

      Well, the $600 billion question, and maybe just to remind folks what, what that is, I mean, it's basically a very simple equation that says if we invest... And this was 2024 when I wrote this, if you invest $150 billion in NVIDIA chips, that's about $300 billion of data center investments, and to pay that back, the person using the compute needs to earn a 50% gross margin, so there's about $600 billion of revenue that needs to get generated. If you redo that analysis in the summer of 2025, it's about $840 billion. So it's, it's grown, but it hasn't grown dramatically. And so the question behind the question was, is the customer's customer healthy? We know that the customer is healthy. We know that people are buying all these data centers. We know that people are building these data centers. We know that those stocks have all gone up. We can see that. But is the customer's customer healthy? Is there actually an end user for this compute? I don't think that's been answered, and I think that the, uh, the question last year which was the valid question was, if everyone's spending all this money, it hasn't showed up yet, because e- people haven't put the shovel in the ground yet. I literally wrote a, a piece last summer called "AI: Is Shovel Ready?" You know, the shovel is gonna start hitting the ground. And so now the shovel is hitting the ground. We're mid-construction on a lot of these projects. One of the predictions I made last year, in addition to saying it was gonna be the year of the data center in 2025, I b- I said, "Hey, we're gonna have these construction delays. We're gonna have issues now in building out these data centers," and the information has done a very good job of reporting on this, but I think we're at the beginning now of seeing some of that play out as well.

   8. HSHarry Stebbings4:39

      Are we gonna see a mass proliferation of delays on data center construction, do you think?

   9. DCDavid Cahn4:44

      I think we're gonna see variability. One thing I'm always interested in as an investor is like, there's winners and there's losers and there's variability, and I'm very skeptical when anyo- and whenever anyone tells me, like, everybody is gonna win or everybody is gonna lose or everyone is gonna do anything. Like, there's always variability. Imagine a race. You have a track race. Like, there's somebody in the front, and there's somebody behind, and someone's faster than the other person. And so I think with data center construction, one of my core perspectives that I've been developing over the last 18 months of writing about this is that construction itself is gonna be a mo- the ability to build things is hard, and I think we underestimate that, and I think we continue to underestimate that 'cause we sort of say, "Oh, well, it's fine. Like, everyone's gonna do it. The timeline is two years." Okay, but like-... there's a lot of (laughs) complexity that goes into that. And by the way, the complexity compounds when everybody is doing the exact same thing at the exact same time, and everyone is trying to buy from the same vendors. And I've written a lot about the AI supply chain for that reason, because you really need to care about not only, okay, META and Google are both building a data center, but who's the guy that they're calling and who's the guy that he's calling? And you gotta follow it all the way down the supply chain to get to the core of really what's going on.

   10. HSHarry Stebbings5:48

       There's so many things I want to unpack within those. I do want to go to, what did you not predict or foresee that did play out that you were surprised by?

   11. DCDavid Cahn5:57

       I think there were two big misses, uh, last year. I think the first big miss was the- these, like, big talent acquisitions. I mean, I think that if you had asked me the probability a year ago that, you know, if you're a 25-year-old recent grad from an elite university who is perceived to be an AI expert, you can get a $5,000-100,000,000 pay package right now. And if you are a brand name that everyone recognizes your name, you can get a billion-dollar pay package right now for a single individual. I totally did not see that coming, and I think had you asked me a year ago to predict that, I would have said you were crazy. So sometimes, I, I do think the beauty of AI is, like, reality is stranger than fiction, and a lot of crazy things happen. The second thing I-

   12. HSHarry Stebbings6:34

       Do you think ... B- before we move to the second-

   13. DCDavid Cahn6:35

       Yeah.

   14. HSHarry Stebbings6:35

       ... do you think those scaled pay packages are justified?

   15. DCDavid Cahn6:39

       I think they're symbolic of this sort of desperation in the ecosystem, where it's like, we need to eke out progress. We need to prove that all these investments are worth it. And I think there's this logic that gets really abused (laughs) in the venture world and i- in the tech world, which is like, hey, if I increase the probability of making a trillion dollars by 1%, that's worth ton of money, right? That's worth $10,000,000,000. And sure that's true, but it's very easy to overestimate the 1%. Is it 1%? Is it a hundredth of 1%? Is it a thousandth of 1%? Is it a ten-thousandth of 1%? Our brains are very bad at reasoning a- at that scale of number. And so, I think to the extent that you believe that hiring this very impressive researcher increases the probability you win by 1%, I totally can see why you would justify a billion-dollar pay package for an individual. That said, I think we are psychologically biased to overestimate what that percent contribution is, and it may be the case that there's these broader macro variables, which we'll talk about, I'm sure, later in this discussion. There's these broader macro variables that are actually driving progress in AI that are, uh, that are not a single individual can change.

   16. HSHarry Stebbings7:40

       I'm very upset looking at these pay packages that my mother didn't push me towards a more-

   17. DCDavid Cahn7:44

       (laughs)

   18. HSHarry Stebbings7:44

       ... engineering-heavy, uh, degree.

   19. DCDavid Cahn7:45

       Doesn't everyone feel that way? I think that's, like, probably the universal (laughs) reaction to seeing these packages, maybe.

   20. HSHarry Stebbings7:50

       I, I ... My mom, you should have done better. Bad parenting. Uh, you en- encouraged me to do English. Really?

   21. DCDavid Cahn7:56

       (laughs)

   22. HSHarry Stebbings7:57

       Um, come on. Um, (laughs) yeah, war and peace doesn't quite make it, does it, when you're getting paid three and a half billion by Zuck? Uh, what was the second?

   23. DCDavid Cahn8:05

       I think the second one ... You know, one thing we talked about on the podcast last year, I predicted that META was gonna do really well. And I think that prediction was clearly false in a 12-month time horizon. Um, I thought that the vertical integration that META had was gonna be an advantage, and I think that META, you know, uh, these 100 million packages are coming in large part from META because they haven't performed as well as they thought they were going to. The reason I thought META would do well is that it was vertically integrated and founder-run. And I, I sort of continue to believe that in the fullness of time, it is possible, and I think the dramatic actions that Zuck is taking represent this. It is possible that I will be proven right in a longer time horizon, which is to say that Zuck's gonna fix the problem. It's amazing what founders can do. He's so focused on this. It's, he's spending all of his time on it. But I think if you look back a year ago at the prediction that META would do well, I think you would say, "Wrong."

   24. HSHarry Stebbings8:53

       Have you changed from a buy to a sell on META?

   25. DCDavid Cahn8:56

       I think it's the dramatic action that Zuck's taking represents just how deeply invested in, in this he is. And I think it also shows us what founder CEOs can do and why founder CEOs are different than non-founder CEOs. I mean, there's all these studies (laughs) of, like, if you just invest in the basket of founder CEOs, you will outperform the basket of non-founder CEOs. And I think what Zuck is doing represents that. And so, I remain optimistic about META long term.

   26. HSHarry Stebbings9:19

       You said about the vertical integration there being part of, like, your thesis. I, I totally agree with you and was probably shaped by hearing you, to be quite honest, David. You said to me data center and model teams need to be coupled, kind of going to the vertical integration element. Do you stand by that? How do you think about that when hearing that today? And does OpenAI and Anthropic not having that vertical integration challenge that?

   27. DCDavid Cahn9:43

       Well, I think the simple version would be OpenAI and Anthropic are now steel servers in power companies. And that's, like, a big change that's happened in the last 12 months. And so, I actually think we're at, you know, in many ways, OpenAI and Anthropic are becoming more and more vertically integrated every day. You're seeing a lot of announcements around them developing their own chips, uh, their work... You know, every day you hear Sam Altman talking about gigawatts of power and procuring his own power. And so, I think you will s- continue to see the big labs moving vertically down the supply chain, and that's been one of the biggest trends over the last 12 months.

   28. HSHarry Stebbings10:14

       Do you think we will continue to see that? We saw Poolside recently announce a two-gigawatt data center that they're building out in conjunction with CoreWeave. Do we think all model providers will need to be vertically integrated in this way?

   29. DCDavid Cahn10:26

       I think that competitive pressures will push all of the model providers to spend more time on this and to have teams focused on this. So, I think the answer is yes. I do think that this is a trend that is gonna be durable.

3. 10:36 – 14:42

   Are We In an AI Bubble?

   1. DCDavid Cahn10:36

   2. HSHarry Stebbings10:36

      When we think about where we are today, everyone says, "Bubble." Uh, you've heard it, I've heard it. It's on my TikTok. Do you think we're in an AI bubble?

   3. DCDavid Cahn10:47

      I do think we're in an AI bubble. I also think, to your point, a year ago when we had our last conversation, it was a very contrarian thing to believe that we were in an AI bubble. Today, it's a very consensus thing to believe we're in an AI bubble. I mean, Sam Altman, Vinod Kosala, Jeff Bezos, like, some of the biggest AI bulls have now come out and basically said, "Hey, we're in a bubble of some s- somew- some sort of the other," and each has their own perspective on exactly how that's gonna manifest. So, I think right now, the bubble conversation has sort of reached kind of t- full consensus. The thing that I think is more interesting is...... who's gonna survive the bubble? What's gonna come next? And so I think there's two components to that. Number one, who are the winners and who are the losers? If you remember from the dot-com, a lot of companies from the '90s still did well. Amazon still became an amazing company after the dot-com bubble. So, I think there's an opportunity for winners to continue to do well after the bubble. And I think the second thing that's really interesting is just timelines, right? Like, a lot of us might... uh, you know, I've always said, like, my core belief is that in 50 years when you and I are 80 years old, AI is gonna have completely changed the world. It's gonna dramatically reshape everything about society. And so if you take that time horizon and you say, okay, AI is this tremendous, tremendous technology innovation. It's the most important thing that's gonna happen in our lifetimes. Probably, it's gonna be among the most important thing that's ever happened in human history and the history (laughs) of this planet, right? So, it is this amazing thing, and yet the market is implying some probability that all of this is gonna happen in such a short time horizon with a very specific chipset and all of this stuff. And so I think unpacking the tension between AI as a long-term winning trend and a long-term generational change and a short-term market cycle that will incinerate capital, I think that's the second kinda area that I think is, is really interesting.

   4. HSHarry Stebbings12:27

      How do you balance that being an investor today, David? Like, you play the game on the field, the Bill Gurley quote, but then also the awareness of the long-term impact that will come over multi-decades.

   5. DCDavid Cahn12:37

      I think it's tricky. Um, and y- I think the one benefit I have is I've been investing in AI for about eight years, and I... so I've been able to... you know, for me, this is not like, hey, this is like, a 12-month thing where you're, like, running and have this FOMO to get into AI. I started investing in AI in Weights & Biases Series A, when everyone said deep learning was going to be tiny. It was a year after the Transformer paper came out, and everyone said, "Deep learning is a tiny market. Why would you invest in this company?" And of course, they had a, a really nice exit to, uh, to Corive recently. I invested in RunwayML when Stable Diffusion hadn't even been born yet, and everyone was saying, "Oh, Transformers is the only way." And of course, Stable Diffusion introduced a new model architecture. And I invested in Hugging Face, which I still remember the first meeting I ever had with Clem. It was the f- you know, he had launched this Transformers library. It's funny, now Transformer's on the tip of everyone's tongue. But that time, NLP, it was NLP by the way, it wasn't AI at that time, and he had this amazing Transformer library, and for folks who are steeped in AI, it was a successor to BERT and this old school of NLP models. So, I just say that to say that, um, I think when you take a long enough time horizon in AI, over the last eight years, you have more opportunity to find investment opportunities. It's not about finding 10 investment opportunities, at least for me. I don't need to find 10 investment opportunities this year. I'd like to find one or two investment opportunities a year that I really love. This year, I've invested in Clay, which I think is an amazing application layer company we can talk about. I invested in JuiceBox, which is building an AI recruiter that has tremendous love. And so I think you can find exceptional AI companies that I believe will do really well over the long time horizon and will continue to succeed for decades and decades to come. And one thing I ask myself before I make every investment is, is this company gonna succeed in spite of market volatility? If your... the only way that your company is gonna succeed is that it can raise infinite capital in a cheap capital market, that's very difficult. If you have real customer love and you've built something that people absolutely need, you're gonna be able to navigate through any market environment. And by the way, we've kind of seen that now with all of these 2021 companies navigating that environment. Some of them came out really strong on the other side. Look at Databricks, 60 billion, now 100 billion valuation. So, you can come out the other side of market cycles if you have compelling product-market fit, a great team, a great founder.

4. 14:42 – 16:09

   Winners and Losers in a World of AI

   1. DCDavid Cahn14:42

   2. HSHarry Stebbings14:42

      So, David, when we play out your question there of the winners and the losers, just so I understand that, who do you think the winners and the losers will be when we look back on this last 12 to 18 months?

   3. DCDavid Cahn14:52

      I've had a very simple framework for this. It's actually, I think, probably the first thing I ever published in AI, in AI's $200 million question way back when in 2023, and, um, the framework is this. Consumers of compute benefit from a bubble because if we overproduce compute, prices go down, your COGS goes down, and your gross margin goes up. So, I've had the view that you want to invest in consumers of compute. Producers of compute, imagine you're producing any commodity asset. If other people produce a lot of that commodity asset, it doesn't matter. It has nothing to do with you. You might be running the best operation possible. You might be an amazing businessperson. But if everybody else starts producing the same commodity asset, prices go down. And so it's very hard to control your destiny in commodity businesses. By the way, this is why commodity businesses tend to trade cyclically and tend to trade at lower multiples than non-commodity businesses. So, I think if you're a producer of compute, you're fundamentally in a commodity business, just like an oil company is in a commodity business, and that is gonna trade a different way and that is gonna have more cyclicality than, than if you're in a non-commodity business consuming the commodity, consuming the energy, and producing intelligence on top of that. And so I think if you're consuming this raw resource, which is power, and you're producing intelligence and doing something that people love with that intelligence, those are the businesses that are gonna do well on the other side of this

5. 16:09 – 22:22

   The Role of Big Tech and Monopolies

   1. DCDavid Cahn16:09

      market cycle.

   2. HSHarry Stebbings16:10

      Aren't three of the best businesses not-commodity businesses in the form of Google Cloud, AWS, and Azure?

   3. DCDavid Cahn16:17

      I love this question, so let's talk about it. I think it's really interesting. One thing I've written a lot about, and y- y- you and I have talked about this, is, like, game theory and these big companies. And one of my core beliefs, or one of the things that I think is underestimated in the market, is that we're living in an anomalous monopoly era. And it's funny because there's so many comparisons to Industrial Revolution, and in some ways, we're living in this new Gilded Age. And we have these seven companies, and they represent 40% of the S&P 500, which is just mind-blowing-

   4. HSHarry Stebbings16:43

      Mm-hmm.

   5. DCDavid Cahn16:43

      ... and have these amazing monopolistic businesses, and, uh, and these businesses are cash cows. And, um, and I think people extrapolate from that, and they say, "Oh, all businesses are monopolistic." I think people have a mental model that implies too much monopoly and not enough commodity, and what I think people underestimate about the big tech companies is that when the big tech companies were founded, when Google was founded, nobody thought it was gonna be a monopoly. Think about YouTube selling for a billion dollars. I mean, that would be crazy if you had known how big all of this was gonna be. So, nobody knew that Google was going to be monopolistic. And you can build monopolies when they're hiding in plain sight. Nobody can see them, and so you build this monopoly and you don't have that much competition. AWS is the same. You mentioned AWS. Nobody knew that the cloud was gonna be this tremendous opportunity when AWS started doing this. And to their credit, that's why they have the biggest market share in the cloud business.And that's been very durable for them. And so, I think when nobody sees the monopoly, you can build a monopoly, and then you can extract margins on the other side. But AI is so different. Everybody knows that AI is gonna be big. Like, this is, I think the irony of the AI, is that everybody knows AI is gonna be massive. But if everybody knows something's gonna be massive, then everybody builds companies. And if everyone builds companies, there's tremendous competition. And so I think the difference between the AI era and the Big Tech era, and it makes sense why everyone is over-indexing or over-training on the Big Tech era, 'cause that's the era we live in. But the difference is that these monopolies are not hiding in plain sight. We all now know that if you build an amazing tech company, it can be worth a trillion dollars. In 2000, if you told people that they could have a trillion dollar tech company, they would have laughed you out the room. And so I think the market environment in which these companies are getting built is dramatically different, and monopoly profits are unlikely to exist. And by the way, final point on this, that's good for us. That's like good for everybody. Like, we shouldn't want monopolies to exist. Monopolies are the- are- are bad for the consumer. The consumer wants to get things for free, and the consumer wants to get things for the cost of capital, and I think that to the extent that there are not monopolies in AI, that's much better for how AI is gonna evolve in a healthy way than if it evolved in- in sort of a monopolistic direction.

   6. HSHarry Stebbings18:46

      You said about kind of consumers of compute will win. I- I like that. But respectfully, it feels relatively accepted in venture ecosystems, for sure, in a way that your bets before weren't. Weights & Biases wasn't, Runway wasn't, Hugging Face was kind of a- a kind of weird community play at a point. What do you think is obvious to you that is not obvious to the rest of the community today?

   7. DCDavid Cahn19:09

      When I first started saying this 18 months ago, it was definitely not a consensus, and so one thing that is tricky in the business of ideas is that as soon as the idea becomes accepted, it was always obvious, but in the moment-

   8. HSHarry Stebbings19:19

      (laughs)

   9. DCDavid Cahn19:19

      ... where you propose a contrarian idea, you know, everyone- everyone kind of criticizes it. So, I do think it's been interesting to see, um, see the change. And then, by the way, the people who had the wrong opinion very quickly changed their opinion such that they were- they weren't actually wrong. Um, and so anyways, I think (laughs) the- the idea game is a- is a tricky one. Um, I think that... A- and the second thing I would say to that is, while people say they believe this, and you and I talked about this on the podcast last year, you probably remember this, everyone says they believe this, and then you look at these pitch book charts where it's like, where's the dollars going? And I think probably 80% plus of the dollars in AI are still going to producers of compute, not consumers of compute. So, I do think you're right that it's an accepted narrative, but the producers of compute consume so much more capital than consumers of compute that if you were s- in a capital deployment strategy and you're trying to deploy as much capital as possible, you have to invest in the producers of compute. And I think that's one of the dangerous things in investing, which is that you have this- there's this almost, like, incentive to invest in people who consume more capital because they're calling you every day, and the people who con- don't consume capital don't want to raise capital, and I think some of the best investments are those companies that don't want to raise capital. When Sequoia invested in Zoom, they didn't want to raise capital, right? They were profitable, they were doing really well. Those are the businesses where I think, as an investor, you really have to focus your time on.

   10. HSHarry Stebbings20:32

       I spoke to Sonia on your team beforehand, and she gave me a fantastic question. She said, "If this is a game theoretic bubble, is there a coordinating mechanism for the spending to stop and the bubble to pop?"

   11. DCDavid Cahn20:44

       You know I love game theory, so... I mean, my- my basic framework on AI, and this is actually kind of how I write all these pieces, is there's like 10 players around this big chess board, and they're extremely powerful, and each of their moves affects the other people's moves, so it's kind of recursive. And- and so you sort of have to think first order, second order, third order, how do- how does my move affect other people's moves? And these are very sophisticated players doing this. And so one- I- what- my- the simple answer to your question is it's- it's not coordinated. That's the beauty of the invisible hand. That's the beauty of people's incentives. These are big companies that are acting out these incentives. And so I think until the incentives change, the behavior is not gonna change, and so there is no coordinating mechanism. I- I do think that's one of the surpris- it's always the surprising fact of capitalism. Like, everyone wants to believe that everything is kind of coordinated. It's easier for our brains to grok everything being coordinated, but I actually think it's- it's pretty uncoordinated and incentive-driven.

   12. HSHarry Stebbings21:34

       If we think about, you said earlier, it is definitely a bubble and we're seeing this consensus across the different visionaries in our ecosystem. If it's a bubble, does it pop or does it deflate, and how do you expect that to play out?

   13. DCDavid Cahn21:48

       So, I'm a student of Nassim Taleb, and I will lean on Nassim Taleb's sort of- he's a f- hedge fund investor and philosopher, and he's written Fooled by Randomness, Antifragile, Black Swan. I think these are books that a lot of folks will be familiar with and- and really influential books in the investing world. And his philosophy, and- and he says this in Antifragile, is, you know, it's really hard to know if a building is gonna fall down, but you can see when it's wobbly. And so you can't really predict when the wobbly building falls, but you can notice the fragility. And so I think my perspective on AI right now is, uh, you can see the fragility. Everybody can see the fragility.

6. 22:22 – 34:26

   Breaking Down Circular Deals in AI: The Truth No One Sees?

   1. DCDavid Cahn22:22

   2. HSHarry Stebbings22:22

      Can I ask you what specifically makes you say you can see the fragility?

   3. DCDavid Cahn22:26

      Well, I- the circular deals. I think the circular deals dynamic is probably... When I think about why did- why did this AI bubble narrative go from contrarian a year ago to consensus today, I think the main thing driving the consensus is these circular deals and the Big Tech company dynamics. Let me- let me unpack that.

   4. HSHarry Stebbings22:43

      Mm-hmm.

   5. DCDavid Cahn22:43

      A year ago, hyperscalers were holding up the AI ecosystem, and everybody felt very comfortable with that because everyone knew that these were very robust businesses. Microsoft and Amazon specifically were driving the vast majority of the AI CapEx growth, and they were explicitly saying, "Hey, we're gonna buy out your generator capacity for five years. We're gonna sign a 20-year lease on this data center, and we'll back it up with our credit." So, they were basically putting themselves in front of all the risk. And the way I thought about it a year ago and wrote about it a year ago is like, they're almost grabbing the hot po- d- demand hot potato and saying like, "It's- it's ours. Don't worry about it. We got this covered."A year later, Microsoft and Amazon have really stepped back. And this started, a- and again, The Information has done a really nice job reporting on this. This started in, uh, the beginning of the year. There was this big, uh, public announcement or, or leak or whatever you want to call it, where, uh, Microsoft walked away from two data centers. And it sent a message to the market, like, "Hey, we're not stepping up. We're not going to take all the risk on everybody else's behalf. We're not going to be this, this, this risk absorber in the ecosystem anymore." And then what happened later this year is Oracle obviously stepped up and took on a huge amount of the compute demand, and Cori has really stepped up and taken on a huge amount of the compute demand. And so you have this shift from Microsoft and Amazon to Oracle and Cori. And then the second order effect of that is that Oracle and Cori are a lot smaller than Microsoft and Amazon. They simply can't absorb as much risk as Microsoft and Amazon could. And so the chip companies are now stepping up and saying, "Okay, we'll absorb some of the risk. We'll put in the capital to finance this buildout," where the demand on the other side is not so clear because, of course, the chip companies also get to book this as revenue. So their, their cost of capital is very low. One might even say their cost of capital is negative in some of these deals. And so it's the, it's the cheapest capital available. And so moving from s- you know, expensive capital from these big tech companies to cheaper capital from the chip companies themselves who get to benefit from circularity, um, I think that's probably been the biggest change in the last 12 months in AI. And I think that's something a lot of people have observed. It's, it's fairly, you know, obvious. And so that, I think, has changed a lot of people's minds.

   6. HSHarry Stebbings24:48

      Do you think these deals are priming the pump, so to speak?

   7. DCDavid Cahn24:51

      I think all of these deals now are priming the pump. I mean, you basically announce the deal, they're 10 or 20% funded, and then you have to go raise capital to, to, to fund the rest of it. And so, you know-

   8. HSHarry Stebbings25:00

      Mm-hmm.

   9. DCDavid Cahn25:00

      ... everyone announces these deals in gigawatts, not dollars anymore. And I think most people don't know how many dollars a gigawatt is. And so the rough math is, you know, a gigawatt is $40 billion to, to build out. Jensen says it's 50 or 60 if you use, uh, the next generation Vera Rubin chips, so let's say it's somewhere between 40 and 60 billion. So 800 gigawatts of power buildout, which is what people are talking about now, that's ai- that would be AI's $8 trillion question. And then 250 gigawatts of power is AI's $20 trillion question. So we've totally upped the ante, and the magnitude is just, i- i- it's just much, much bigger. But of course, that's not funded. Um, and so I think the funding for these deals is, is gonna be an important thing that has to play out.

   10. HSHarry Stebbings25:41

       How do you read them? When I hear you speak now, I, I feel very concerned. Like, I think, is there even the capital supply in the world for these? You know, we've heard about Sam Altman and the trillion dollars that he needs and r- requiring the same energy as Japan, and you're actually looking at that going, "Well, not even the sovereigns have enough money for that actually."

   11. DCDavid Cahn25:59

       Well, we're living through this amazing moment, and I do think it's precarious. We're living through this amazing moment where, like, the entire capital market is just AI, right? I mean, 40% of the S&P 500 is these big tech companies. They're all basically trading on AI. Uh, private capital is all targeted at AI. And so I do think the world's capital machine is directed in a single direction. I think the risk is that it's all focused on a very constrained period of time. I actually think in the fullness of time, it's not that risky. Like, these things are gonna play out. We're gonna get these amazing... AI is gonna be amazing. We're gonna get these huge technological breakthroughs. Tremendous revenue is gonna get created. It's gonna be a dr- big driver of the economy. The problem is, and, and the simple way to think about it is, it's all B100s and H100s, and what if it actually takes three years and it's the Rubin chips that get us there or it's the Feynman chips that get us there, which is the 2028 chip, right? So I think, again, it comes back to where we started, which is the physicality of AI. You can't just say like, "Oh, I'm upgrade my chip. Great. Snap my fingers, I've upgraded my chip." No, you have a, a giant warehouse sitting with these chips, and they might be legacy chips, and maybe it's gonna take us 10 years to get there instead of two years to get there. And I think that is kind of the risk that the financial ecosystem is taking on. Whereas as an AI investor and an AI believer, I'm like, "We actually just need to spread that risk over a longer period of time and a lo- and a greater number of, of bets."

   12. HSHarry Stebbings27:17

       Oracle is one of the biggest players that we've seen enter the market, as you mentioned there. When you look at their, like, debt-to-equity ratio, traditionally considered very, very high, do you not think they're out over their skis?

   13. DCDavid Cahn27:28

       I think that one narrative that I have been, uh, thinking about a lot is this narrative that I think a lot of the media has also been painting of like, "Hey, debt is gonna unwind the AI bubble," which is to say, "A lot of these AI investments are debt-funded, and the problem with credit is that credit unwinds, and then when you have a credit unwind, a lot of bad things happen." I just think that's not the way it's gonna play out, which is maybe surprising. Um, I think that the reason people are so anchored to this sort of debt, debt narrative is that 2008 was a debt credit unwind and people understand how messy credit unwinds are. I actually think that what's interesting about this AI buildout is that for the most part, and let's put Oracle aside which maybe has some debt, but for the most part the AI buildout today has been equity funded and cash funded. And so I think it's actually, i- i- is, you know, every, every bubble looks different and every unwind looks different, and I think we always sort of over-anchor on the, the lessons of the past. What I think is gonna be interesting if, to the extent that, that the bubble unwinds at some point, it's gonna mean equity unwind. And then what that looks like is 40% of the S&P 500 is basically a bet on AI, and so to the extent that the bet unwinds, stock prices go down. And what's different this time, again versus 2008, is more Americans, uh, you know, a greater percentage of Americans' net worth is equities than I think ever before in history. And so people are gonna feel this in the form of their equity portfolio going down p- more likely than some credit unwind where the banks get affected and all of that stuff.

   14. HSHarry Stebbings28:49

       Are you as concerned as I am by the concentration of value in Mag 7? It's not a... And again, if I'm pushing you on company specifics, dude, I mean, I, I... No, really, I'm not a journalist in any way, like I-

   15. DCDavid Cahn28:58

       It's all good.

   16. HSHarry Stebbings28:59

       I have zero, I have zero desire-

   17. DCDavid Cahn29:00

       (laughs)

   18. HSHarry Stebbings29:00

       ... to get a clickbait answer. But, like, I look at the concentration of value in Mag 7 as a, as a class or cohort, and I am worried.

   19. DCDavid Cahn29:09

       Yeah, y- I was sitting down yesterday with, um, Sandy Noren, who's the author of this book, The Engines That Move Markets, which is one of the all-time great tech investing books, and we were talking about AI, and we were talking about markets. And he sort of made this comparison to Japan in the '90s, where basically if you didn't inv- if your portfolio was not leveraged to Japan in the '90s, then you were, like, the best performing fund in the '90s. And that it was, I think he said, and this was, like, really surprised me, he said that, uh, Japan was basically 43% of the equity market, and the US was 41%. And so it was really, really a huge percentage of the market, right? And that really unwound. And so, I think you have a similar dynamic here where the Mag 7 are just a humongous portion of the market. Now these companies are great. They have cash machines. Like, they're gonna do fine, but I do think we should be concerned that these companies represent such a huge fraction of the market and that any change in the AI narrative really affects them.

   20. HSHarry Stebbings29:58

       I, I want to discuss, you mentioned earlier in the conversation, and we mentioned the concentration of value in Mag 7. A lot of that's pr- predicated around the belief that it will impact GDPL- GDP meaningfully, and we touched on it earlier. Masa said that he thinks that we'll see 5% GDP impact. How do you think about and respond to the magnitude of which we will see AI impact GDP and productivity levels?

   21. DCDavid Cahn30:24

       So, I think Masa makes an interesting point here, and I actually agree with him fundamentally that AI is gonna affect 5% of GDP. Probably where I disagree with Masa, so I think he, he used the $9 trillion. I think that's the number he used. It's gonna-

   22. HSHarry Stebbings30:35

       Yeah.

   23. DCDavid Cahn30:35

       ... disrupt 9 trillion of GDP. And then he says his next assumption is it's gonna fif- that's gonna, it's gonna be a 50% profit margin, and then it's gonna be $4 trillion of pr- economic profit. And I think, so I agree with him, it's gonna affect 5% of GDP, maybe more in the fullness of time, um, but I think this comes back to the point we were discussing earlier where people, uh, over e- estimate the monopolistic nature of businesses and that we're living in this sort of unique Gilded Age monopolistic era and that that is, is not the steady state of business. And I was reading, I found this McKinsey report recently which said that if you look at total global GDP, 1% of global GDP is economic profit above the cost of capital, which I think is surprising and I think, again, confirms this intuition that I think some people, that, that I think is important, which is for the most part, GDP accrues to regular people, working people who get wages and salaries, and, um, it is very hard to sustain an economic profit above your cost of capital. And again, to, to, to moralize for a second, like, that's a good thing. Uh, I do think that's really good, and I hope that the economic benefits of AI accrue to everybody and not just a few companies.

   24. HSHarry Stebbings31:42

       In terms of overestimations, you know, I, I was just chatting with Rory, uh, O'Driscoll from Scale and Jason Lemkin who we have our weekly show, and they actually said that the biggest problem with today is we're seeing this overestimation of demand. They were specifically talking about legal, um, where every law firm is looking for an AI provider today because they've been told, "Look for an AI provider." That will not be the case next year and the year after, and so it is a atypical market cycle where 100% of market is looking for a new provider or a provider where normally it would only have been 5%. Do you think that's a fair description?

   25. DCDavid Cahn32:21

       I think that we are, I think there's a number of things that are over, being overestimated. I think the most important one is the timeline, and, I mean, you've probably seen there's a lot of commentary now in the last few days about this, like, AGI timeline getting pushed out, and, um, and p- you know, this is something I've been talking about for the last, like, four months and, because a lot of the leading indicators were there in June, July, but this did change over the summer, so it makes sense why everyone's talking about this right now, which is in June or July, Andrej Karpathy at Y Combinator said, "Hey, we're in for the decade of agents as opposed to AGI in 2027." And, uh, a few weeks ago, Richard Sutton was on the Dwarkesh podcast and basically explained why, and, and Dwarkesh, I think, has been doing a good job of fleshing out why the current technology paradigm is not enough potentially to get us to AGI. Um, and so, and then Sam Altman came out, I think also in June or July and said, "Hey, it's gonna be a more gentle singularity." I've actually been surprised by how, you know, uh, gradual the change has been as opposed to being sort of this, this crazy change. And so for me, there's this contrast between what I, I think of as, like, the lunchroom conversation at these big labs. Like, you have these 25-year-olds sitting around lunch being like, "AGI's 100 days away. No, it's 200 days away. No, it's 300 days away," and, like, the highest status person is the person who says, "It's 100 days away" because they're the most aggressive. And, and then, but you contrast that against, like, the true thought leaders and, and godfathers of AI, the people who really invented this category, people like Richard Sutton, people like Andrej Karpathy, people like I- Ilya Sutskever who said in December that pre-training is dead, and those people think, "Hey, the timeline's actually, like, 20 years, 30 years, et cetera." And so I think that contrast is probably the biggest thing that's being underestimated, and I think the irony of that is that it's actually the people who are the forward-thinking leaders who sort of led us down this path, like, the path we're on was invented by these people who are raising the most concern or saying the timeline is longest, and it's the people who've been in AI the shortest who I think are saying like, "Hey, it's gonna come tomorrow," and I think there's sort of this experience curve of these things are just hard, and they take time. And by the way, just to say this 'cause it's so important, it's like if this happens, it's a cataclysmic event in the history of our species, so it doesn't really matter if it happens in 200 days or 50 years. What matters is

7. 34:26 – 37:53

   Why Kingmaking is BS and VCs Do Not Make or Break Companies

   1. DCDavid Cahn34:26

      that it does happen.

   2. HSHarry Stebbings34:27

      I, I almost feel apologizing 'cause you're so smart and intellectual, and then I'm like, "Yeah, well venture baby." Um, but like king-making is a real thing. Making one person the anointed winner with a large amount of capital distribution and brand a la Harvey is a very real dynamic that we're seeing play out. How do you balance that, the importance of king-making today with the long cycles, the decade plus that we're talking about there?

   3. DCDavid Cahn34:56

      I don't believe in king-making, and that's maybe a controversial thing to say. I think one of the lessons, you know, you'd think like, oh, Sequoia should be able to king-make companies and, like, that's so great, and that would be... By the way, if that was true, it would be really economically valuable for our LPs if that was true, and I don't think that we think that's the case. Uh, and I think if anything, some of the hardest learned lessons in this business are, like, you think that your capital is gonna change the business. Uh, it's not.It's not. Fundamentally, the founder has to be amazing, the idea has to be amazing, product/market fit has be- to be amazing. Maybe we can help them navigate a few difficult decisions along the way, and we like to think of ourselves as company builders. But I think the lesson that punches you in the stomach in venture is you can't make a company succeed. The company has to already be successful. And then I think the second order effect of that is, like, you should be humble because the company succeeded not because of you, the company succeeded because of the founder, and maybe you helped a little bit. But, um, you can't make companies succeed as a, as a venture capitalist, and I think that, um, ego gets in the way where people think they can, and, and I just don't think they can.

   4. HSHarry Stebbings35:52

      So you don't think in a market like Profound that Sequoia and the subsequent Quick Round has helped them significantly get great talent, get great customers, and get subsequent funding, which has then widened the moat between them and the plethora of other people? I- I'm sorry, I, I love you, but I respectfully disagree.

   5. DCDavid Cahn36:14

      I think that there are flywheel dynamics, for sure, in venture, and so I'm not saying that having... I, I think having Sequoia on your cap table makes your company more successful-

   6. HSHarry Stebbings36:21

      Yeah.

   7. DCDavid Cahn36:21

      ... so I'm not saying that having a brand name great VC who's gonna work really hard on your cap table doesn't change the probabilities. I just think it changes the probabilities less meaningfully than people think, on average. And I think that, you know, you should use Profound as an example because I was in the pitch when they came to the IC. The business was ripping. It was an amazing business. They had tons of customers lining up at their door to buy the product. And so, yeah, we're lucky to be in business with them and we're grateful to be in business with them, and I hope that we can shape the journey in some way. And if there's five engineers that join and Sequoia help, can, you know, help, uh, having Sequoia involved help them join, phenomenal. And by the way, I think that's the number one way that companies do benefit from having Sequoia on the cap table is that, is talent and recruiting, and we can talk more about that. And I'm, I'm fascinated by recruiting and recruiting dynamics. So I do think Sequoia helps with that, it especially helps with folks who are more memetic where I think the, the, the brand name really helps. That said, I, I just resist the idea that like, oh, you know, I think this is just something that you learn the hard way in this business, like, "Oh, I'm gonna put 20 million in this business, now it's the Sequoia company in this space and suddenly it's gonna succeed." Like, no, I don't th- it doesn't work that way. We've learned that the hard way, and I think we, in our investment committee conversations, we really resist that, because I think that is how you make mistakes in venture.

   8. HSHarry Stebbings37:29

      It's so funny, I remember when I interviewed Doug and he was like, "People think that, like, 'cause we're Sequoia, everyone just comes and says, 'Oh, here you are, here's my whole deal, you must have it, take it.'" And he's like, "I wish. I would love that. It's not how it works." (laughs) Like, "I have to fight and fight and fight." And I'm like, "Uh, yeah, your biceps are bulging, Doug. I totally believe that you have to fight (laughs) for the, for every deal. It's all good."

8. 37:53 – 39:54

   The Importance of Margins in AI Investments

   1. HSHarry Stebbings37:53

      Um, you mentioned a couple of companies that you work with. The common critique posed to consumers of compute is margins, margin structure, unhealthy margins. Do margins matter today in this entry point of AI, or not?

   2. DCDavid Cahn38:11

      I think they matter, and the companies I've invested in typically have reasonably high margins. Um, that said, I think they matter, they're, they're a directional indicator of how much product you've built on top of the foundation models. They are not absolutely important. I, you know, I remember investing in a company many years ago that had a 30% gross margin, and now it has a 70% gross margin. And so gross margins go up over time. I think one thing, as an investor, that I guess you viscerally experience is that plenty of companies that get critiqued for having low gross margins end up being super healthy businesses in the long run. You know, Snowflake was... One of the big, uh, indicts on Snowflake in the early days was that it had a low gross margin. Obviously, it's a tr- it's a very good business. So I think if you have a real product that delivers a lot of value and there's reasons why, as you get bigger, the cost is gonna go down... And in AI, there's such an obvious reason, which is the cost of compute just keeps coming down every year. (laughs) So i- the trend line is very clear. I think if you build a healthy business, and so I would even go to the extreme, and, and I haven't invested in any of these companies, but I would go to the extreme to say that even some of these companies with 0% gross margins, I can imagine how they're gonna work. Now, the companies I've invested in typically have higher gross margins, um, than that, and, and I think that's an indicative of the amount of product that they've built. At the end of the day, our job is to invest in companies that become really successful, not to be, like, super smart about analyzing them. And so I think sometimes the instinct to criticize a gross margin can get in the way of money making. And, uh, you mentioned Doug. I, I, I sort of... The thing I've learned from Doug or the thing I admire most about Doug is, like, the job is to make money at the end of the day for LPs, for founders, for everybody. We all ha- you know, that's the business that we're in, and so I try to keep that as the, as the goal at the end of the day.

   3. HSHarry Stebbings39:43

      Uh, I, I have something called WWDD, which is what would Doug do? (laughs)

   4. DCDavid Cahn39:50

      (laughs) Yeah.

   5. HSHarry Stebbings39:50

      Which is, like, in a tough situation, I'm like, "Hm, WWDD?"

9. 39:54 – 49:13

   The $0-$100M Revenue Club: Is Triple, Triple, Double, Double Dead?

   1. HSHarry Stebbings39:54

      Um-

   2. DCDavid Cahn39:55

      It seems like a good framework.

   3. HSHarry Stebbings39:56

      ... we mentioned, uh, margins is one. Growth rates is another, that companies are just growing so much faster than we've ever seen before. We... I had Hemant on the show from GC. He said, "Trouble, trouble, double, double, I say go." Like, you know, "Come back when you got something better." Brian Kim said recently, and it caused a lot of furor, like, if 2 million in AR are like, you know, in a 10 days, like, come on. How do you feel about this growth rate on steroids requirement from VCs, and how do you feel? I- is triple, triple, double, double dead?

   4. DCDavid Cahn40:27

      I think of it a- as, I think of it as the zero to 100 club, so I think it's a variation on this, which is the best AI companies right now are going zero to 100 million of revenue very quickly. And I don't think you have to be at 100 million (laughs) in revenue, to be clear, but I think that, uh, as an investor, you wanna believe the company is gonna be one of those companies. And I think companies that are on that trajectory or have crossed that trajectory are companies like Harvey and Open Evidence and, uh, and I think and Clay and Juice Box, and I think these are companies that are kind of on this trajectory of growing really, really fast. Um, the reason why it's important is because, to your point on how there's so much demand right now for AI, the best companies... It is the best indicator we have that you built something really useful. People are... And we, we've talked about this actually a number of times in our partner meetings at Sequoia. You know, you sort of look back at the internet, there weren't that many people on the internet, and so these companies could only grow so fast. Right now, everybody's on the internet-... and everybody wants to buy AI. So if you have something really good, it's going to get adopted really fast, and so I do think, to the point of playing the game on the ground and adapting to what you see in the market, the biggest thing that we've seen in the market is that these companies growing zero to a hundred are the companies that have smashing product/market fit. And so I'm happy to invest in a company with 2 million ARR that is smashing product/market fit. What I would tell you is the companies with smashing mar- product/market fit are growing faster right now. And by the way, they don't always have (laughs) to grow faster. Like, the goal is to invest in something that in 20 years is this amazing public company with billions of dollars of revenue, and that is still the first order thing. But I, I, I think you, um, you know, don't fight the tape. Like, you can't ignore the traction on the ground.

   5. HSHarry Stebbings41:51

      I always say, I don't care how long you take to get to a million in revenue, but I care desperately about how long it takes for you to go from one to 50. (laughs)

   6. DCDavid Cahn41:59

      Yeah, yeah.

   7. HSHarry Stebbings42:00

      I, I r-

   8. DCDavid Cahn42:00

      There's a lot of, there's a lot of data that indicates that that is a very good leading indicator, for what it's worth. The, the data I've looked at suggests that that is a, a historically good algorithm.

   9. HSHarry Stebbings42:08

      You know, one of yours is UiPath, and he's a dear friend of mine, Daniel, and I mean, it took nine years to get to 550K of ARR.

   10. DCDavid Cahn42:19

       I wish I'd invested in him in the first few years. I got to work on the-

   11. HSHarry Stebbings42:21

       So-

   12. DCDavid Cahn42:21

       ... investment when it was later stage. But, I mean, obviously, amazing story and I think one that should inspire people. One thing I try to talk about with founders also is, like, I want to inspire founders (laughs) that it can take a, a long time because Silicon Valley sometimes has this, such a short-term time horizon. And I look at Juicebox, you know, it was a company who started three years ago. The founders, uh, uh, the CEO is 25, uh, the CEO is 22, sorry, he's now 25. The CEO is 22. He had dropped, y- you know, finished Harvard in three years. The CTO dropped out of Dartmouth, he was 19. They took them three years. They were always focused on recruiting. They had an initial music app in college, and they evolved that into the recruiting market, and they spent three years figuring out what their product should be. And now, of course, it's growing really fast, and they're really good founders. And one thing I've learned, and this incentivizes me to invest in companies like this, is people like David and Nishant, the Juicebox founders, who've sort of been through the founder journey, they've been through the pain, they understand how hard product/market fit is, I think in the fullness of time, they are better founders for it, and, uh, those sc- those scar tissue, even though they're really painful, I do think they pay dividends long term. And I think for founders who are listening, who are, like, in year one and things are hard, you know, that's, that's, you know, I, that's painful, and there's nothing that I can say that's gonna make that less painful. But I think there is, like, we would love to invest in you, to yours- as you figure it out, and, and we're super patient. And the- not- most- there's this n- m- false narrative, I think, that, like, all the good companies, they, you know, they raise the seed and then they raise the A and then they raise the B and it's all in 12 months, and the revenue, uh, that's not really how most of these companies work. Clay spent many years ... It's funny 'ca- we've been talking about Juicebox and Clay. Clay spent many years in the wilderness figuring out what their product was gonna be. Sequoia invested at the Series A, uh, in, I think, 2019. The company spent three or four years in the wilderness really figuring it out. I look at Karim, I think the man is, like, enlightened from this experience, like, it's super painful, uh, experience. Uh, Varun ended up joining as a later co-founder. Amazing combination. So the company completely changed from the Series A, and then I led Sequoia's investment at, you know, a bil- a little north of a billion, um, wh- which w- we are doubling down in the growth stage of the company, and obviously the company now has, uh, has continued to rip and has done, has done really well. So I think the, the default narrative of like, oh, I'm gonna start the company and then 12 months later I'm gonna be successful, at least in the case of two of the investments that I'm most excited about, that was definitely not what happened.

   13. HSHarry Stebbings44:31

       The reason you come on the show is 'cause I stalk the shit out of you.

   14. DCDavid Cahn44:34

       (laughs)

   15. HSHarry Stebbings44:34

       I spoke to Varun and David from Juice bo- Varun from Clay and David from Juicebox before. Both said, "I told you, I didn't have one person not respond-"

   16. DCDavid Cahn44:42

       Ah.

   17. HSHarry Stebbings44:42

       "... to my calls or messages about you," which is, like, very, very rare, dude. Like, that's testament to you. You mentioned that like, oh, for founders who, you know, i- i- it's hard and, you know, we don't wanna present this false picture of being easy. Completely true, but we are seeing these very quick successive rounds. Y- you know, if we look at, say, a Relate or a Profound or a ... Do you worry about them? I remember Pat Grady once saying to me that his biggest challenge is that when he does a deal, everyone else wants to put in money at double or triple the price, and that really stuck with me. Do you worry about these very quick successive rounds?

   18. DCDavid Cahn45:18

       I think we try to find the right balance and, uh, to be, to be honest, this is a conversation I have with a lot of founders, right? So this is like a very active conversation. We're all having these conversations all the time, and we're obviously in a market where capital is very abundant and very available, and so I see the argument for why people want to take the capital. I think one lesson we've learned is more capital does not make a company more successful. Capital is a, is fuel, but it, capital does not create the engine. And so I think this is a tension, I think this will always be a tension, and I think this is definitely a tension for companies right now where, and we learned this the hard way in 2021, getting over-capitalized has downsides. I think it leads to ... The biggest downside, in my opinion, is that it leads to this sort of internal perception of like, we're, we're winners. We're so successful. We're so great. And, um, the only thing that makes you a winner is having tremendous product/market fit and having customers who love you. And, um, and so I think that's a tension. Some founders, and I've seen some founders do a great job of this that I've worked with, they, they, they really act like the money is not in the bank account and they really behave diligently and the team size doesn't grow too fast and all of this stuff, but I think that is the exception, not the rule. And I think it's actually the f- not the founders that are i- uh, you should be most worried about, but it's the engineer who joins the company the day after the billion-dollar fundraise with very little revenue. That dynamic is tricky, and I, um, I admire the founders for navigating it. Don't think there's an easy answer. I wish there was. (laughs) I don't think there's an easy yes-no answer, uh, but I think it's a tension we should be talking about, and as company builders, it's something that we need to, uh, we really need to think about.

   19. HSHarry Stebbings46:44

       Speaking of Pat quite a lot, poor guy. It's like an advert for Pat. He taught me something that was really interesting, which was two questions which are a framework for amazing insight from founders. And he said number one is, like, uh, what does everyone think they know that actually they get wrong?

   20. DCDavid Cahn47:03

       Mm.

   21. HSHarry Stebbings47:03

       If we apply that to AI and what we see today, what does everyone think they know that they actually are getting wrong?

   22. DCDavid Cahn47:10

       I guess I would say, and this is a really hard one lesson and it's something I've learned from a lot of my mentors in this industry because I think one of the things I really try to do is learn from people who've been doing this for longer, who are smarter, who are more thoughtful, and, uh, one lesson that I've learned in this business is that anything multiplied by zero is zero.And I think that's one of the really tricky things in investing, which is just to say that market volatility doesn't matter, in the long run, if you have a great business. But if you overextend yourself, and then some crash happens and you go bankrupt, you're bankrupt, right? Like, there's no way out of that. And so, and I think that there's sort of this sense, I heard this phrase recently, momentum has its own reality. And I think there's this sense of everyone is living in this, like, reality distortion field of momentum, and, um, I think of it almost like this boomerang, like, you know, this slingshot. You, like, pull the slingshot back, and then, you know, you release the, the, the thing, and then it sort of- it has its own momentum after that. And, and that's sort of a fundamental law of physics. Things in motion stay in motion, things at rest stay at rest. And so, I think the thing that kinda people think- uh, feel so confident in is this, like, reality distortion field that comes from momentum, and, uh, when that reality distortion field goes away, uh, you need to survive that. And I think one thing that I hope that I can be to my founders is a partner and then I- I, you know, they're- they'll listen to me 10% of the time, and that's fine, but a partner in, um, you know, making sure that we survive those moments and navigate those moments well and position ourselves well against that. And I think that the most prudent of investors, or like the most sober of investors can actually be really helpful. Your job as a founder is to be maximally aggressive, and you should do that, and then the, the found- the, the investor should hopefully be giving some advice, helping think these things through, giving some perspectives, uh, you know, understanding the- a, a broader time horizon perspective, a- and a broader data set of companies, and then you sort of navigate to the right- uh, to the right end destination. So I think, um, I don't think people are thinking about this sort of concept of anything multiplied by zero is, is zero because the time horizon is so compressed into this shorter period of time. Um, and, and that's just something that I think

10. 49:13 – 58:29

    Why the Most Important Hire for Startups Today is 23 Year Olds

    1. DCDavid Cahn49:13

       a lot about.

    2. HSHarry Stebbings49:13

       The final one that Pat taught me, and then we'll move to talent, which I do wanna touch on for a quick fire, but he said the other one-

    3. DCDavid Cahn49:19

       Pat is very-

    4. HSHarry Stebbings49:19

       Yeah, is like-

    5. DCDavid Cahn49:19

       Pat is a very, very, very good guy.

    6. HSHarry Stebbings49:21

       What is no one thinking about that everyone should be thinking about? So like for me, uh, what I think is astonishing is like no one is thinking that if you foie gras engineers in terms of the capital that you are stuffing down any of the multi-billion dollar, you may not get an equivalent level of productivity as when they didn't have multiple billions of dollars. Give a nerd billions of dollars, nerd buys five cars and a boat. Nerd not so productive. Like, I'm sorry to be so blunt and direct, but it's the same with companies. (laughs)

    7. DCDavid Cahn49:53

       I think that companies underestimate 23-year-olds and 24-year-olds. I think this is something that people really, really underestimate, and I think this is more true than ever right now in AI. Like I, you know, (laughs) I recently, like I- I meet probably 200 or 300 young recent college grads every year, and the reason I meet them is I wanna recruit them into my companies, a lot of them are founders, uh, this is the population that I learn the most from because I know that my blind spot is gonna be that somebody started using ChatGPT when they were 18 and, and I didn't, and- and so they're gonna have a different perspective, and that's the perspective I need most in my life. In any case, I introduce some of these people to companies, and the company is like, "Well, what's their skillset? Like, why should I hire them?" And, um, you know, I guess, I think this is something that people are not thinking enough about in AI right now, which is ChatGPT's been around for five years. Nobody has more than five years of experience in AI. The playing field is super level. And I think in a changing and dynamic market environment, uh, dyn- dynamism, and slope, and ability to learn are more valuable than ever. And so the thing that inspires me and the thing I spend a lot of time thinking about is, you know, in a JuiceBox, for example, how can we get the very best 23-year-olds in the world working at this company? And that's a big part of my job, and I spend a huge amount of time on that, a huge amount of time on it. I'm there one day a week right now at JuiceBox, just working on this. So how do we get the best people in the world inside of these companies? And I think maybe 10 years ago, in the era of software, um, you know, a senior software engineer, a staff software engineer, they had more experience than a j- than an L3, and, uh, you know, architecture's hard, writing code is hard, and they- they were much better. And so maybe it made sense, there was this old playbook for startups of like, oh, you hire this staff software engineer who kinda knows what they're doing and you don't have to train people. I think that the new playbook for these AI startups is actually gonna be much more about hire the AI generalist, this 23, 24, 25-year-old who's really native in AI, really passionate about it, and I think those are the- the sort of the front lines that are gonna make great companies.

    8. HSHarry Stebbings51:43

       Totally agree and understand that. Do you worry about emotional maturity a little bit? And I don't mean that patronizingly, but Jesus, I mean, I'm 29 now, but when I was 22, 23, I- I- I did some things that I would not do now.

    9. DCDavid Cahn51:56

       I think that hiring always has trade-offs. W- I think one thing I believe more generally speaking, 'cause it's worth saying, is, um, I really believe in trade-offs. I think everybody wants the free lunch thing. When the- when you don't know the trade you're making, the- then the- the negative is hiding from you. There's no such thing as a trade without negatives. There's no such thing as a decision where it's all positive and no negatives. So I- I always talk, and I talk about this a lot at Sequoia actually, it's like hidden risk versus visible risk. And so when you hire a 23-year-old, there's a very visible risk. They're emotionally immature, they don't have any work experience, it's very obvious the negatives that you're taking. When you hire someone who's more experienced, it's li- like less obvious the risk that you're- the- the risk that you're taking. It seems to be lower risk, and every decision is a risk, right? And so maybe the risk that you're taking is that they're not gonna work as hard. Maybe the risk that you're taking is that they're less AI native. There, you know, there's always a risk, and I think people have this tendency to favor the hidden risk. By the way, price is a hidden risk. You don't perceive it as a risk, but it is a risk. Um, and so people prefer hidden risk over visible risk, and I prefer visible risk. I wanna know exactly what risk I'm taking, and then by the way, I'm a huge risk taker. I started investing in AI eight years ago, right? Like, I love risk. So I think it's important to calibrate that like I love risk-taking, but I wanna take visible risks that I know the risk I'm taking, and I think herd behavior and consensus mentality is about hidden risks. The risk is just beneath the surface and you're not paying attention to it. Um, whereas I wanna take risks that I can see, and I think there's a lot of areas. The point I'm trying to make is, um, in the hiring (laughs) dynamic, when you hire a 23-year-old, it's like super obvious why you shouldn't hire them, and yet sometimes that's okay because the reason you should hire them makes up for that more.

    10. HSHarry Stebbings53:27

        Completely agree from the employer side. On the flip side, um, when you think about, like, advice to them, if- if- uh, you were advising your younger sibling on choosing their first job, I- I saw on LinkedIn, you said, "Follow the smartest people a year ahead of you," that moniker of advice may not be relevant anymore, what advice would you give to them?

    11. DCDavid Cahn53:47

        Well, this is, like, the biggest learning. Because I've met with two or 300 young people a year, I have a very big data set, and I think I've probably spent more time than anybody at Sequoia on this specific, you know, thing. And my biggest lesson is that the way that young people choose their career is this, what I call, the memetic algorithm. And the memetic algorithm is, yeah, "What did the people one year ahead of me in school that I thought were the best, what did they go do?" And it's a recursive algorithm, right? So, it's like, "What did the people a year ahead of me do?" But those people chose based on the people a year ahead of them did, and those people chose based on the year ahead of them did. Now, one reaction to that would be negative of, like, "Oh, that's so memetic, they should think for themselves." I actually don't have that perspective. I'm fine with it. I think it's, like, a reasonably good algorithm. When I graduated from college, Palantir was the hottest company to go work for, all the really smart people went to go to work for Palantir. Going to work for Palantir would've been a great life decision at that stage. Uh, before that, you know, in the early 2010s, Google and the big tech companies were the hot place to go work, and I think, you know, those companies were all 10Xs over the, uh, over the 2010s. Some of them, even, I think, 25Xs. So, the- the- it was a good decision to go work at Google in 2010. And so, I don't think the memetic algorithm is inherently broken, and I respect it, and- and I think that people, to your point, a maturity, people are gonna go through a maturity curve, they're not gonna use this algorithm when they're 30s, they're gonna evolve, they're gonna change. And so, I- I sort of have this respect for it. That said, I do think that recursive algorithms break down in the face of dramatic new data, and the dramatic new data is the AI cataclysm. AI has totally changed how the world is gonna work, and it should change your forecasts on the future. And so, the recursive algorithm of, like, "What did the guy a year above me and the person a year above me do?" is actually breaking, because those people didn't have this information. They didn't know that AI was gonna change the world. They didn't understand, uh, gen AI. And so, I think the advice that I try to give young people is, just factor that into your algorithm. Like, you do you. It's like, join the company that you wanna join, go to the place that's gonna make you the happiest, but, you know, factor that in. And then, it's worth at least giving a shout-out to this group of people that I call builders in this, uh, Substack post that I did, which is, builders are people, most people, 90-plus percent of people, their question they're asking when they're choosing a job is, like, "What can I get from this job? What is it gonna enable me to do? Who am I gonna surround myself with? How am I gonna become better?" It's very- it's a very, like, "What do I get out of it" I think there's, like, a 10% group of people, maybe it's 5%, maybe it's 1%, I don't know exactly what the percentage is, but there's this group of people that they're asking the question, "What can I contribute?" And by the way, if you contribute a lot, you generally get to extract a lot, and so I think contribution, this is, again, the beautiful thing about capitalism is, like, when you contribute a lot, I do think that you get rewarded for that. And so, those are the people driving Silicon Valley, like, those are, when you go into a company and you're like, "Why is this company succeeding?" it's those type of people, and those are the type of people who, like, they go from one great startup to another great startup to another great startup. And so, anyways, that distinction between these two groups of people, both valid, no problem with either of them. Like, you gotta respect career is a very personal decision. Um, and so, anyways, uh, depending on what's your- what you're trying to solve for, what- what's gonna grow my career, option one, where can I contribute the most and therefore extract the most, option two, I think there's, uh, a bunch of great opportunities ahead of you, and, um, just factor in the AI variable.

    12. HSHarry Stebbings56:48

        I think one thing that just frustrates me on this topic is, like, the memeticism that continues despite market changes in the UK. And what do I mean by that? Gold- Goldman Sachs investment banking consulting is still whatever people tell you. If you go and speak at universities, which I do once or twice a week now-

    13. DCDavid Cahn57:07

        Wow.

    14. HSHarry Stebbings57:07

        ... everyone still wants to be an investment banker. And so, my, uh, when you were talking, I was thinking, "Well, what does it take to break the memetic chain?" And maybe it's AI and the proliferation of AI in popular culture and media, um, but-

    15. DCDavid Cahn57:21

        I think it's changing. I agree with you, like, it's changing too slowly, and that's why I'm having these conversations and I- I'm- I'm trying to help, and I'm sure y- you are as well in- in these talks that you're doing. I think that, um, one positive that I would say is, I've seen a material change in the last 12 months, which is sort of interesting, 'cause it's not like, I'm not saying the last 24 months, I'm not saying the last 36 months. Like, it took two years after ChatGPT for this to really start flowing through. But I would say, it is, and by the way, a lot of the people I'm talking to are currently investment bankers who wanna get into AI companies, so it is sort of funny that way. I think there's more and more of these high-performing people want to be inside of AI companies, and that's why I think it's sort of a- it is a two-way match. Like, these companies need these people more than ever, but I think these young people can benefit more than ever from being in an AI company. And again, maybe to make the value prop clearer for, like, the young person, right, like, the value prop is, hey, maybe 10 years ago if you joined a startup, and- and people didn't join startups that often 10 years ago, maybe 10 years ago if you joined a startup, like, there's this whole experience curve, you're the junior engineer, there's a lot of people who are smarter than you, and you're gonna have to learn, and it's, like, gonna take five or 10 years to become a really meaningful contributor. That's not really true anymore, right? You're sort of entering at, like, much more parity with everybody else. And so, I think there's good reason why people are making

11. 58:29 – 1:07:33

    The Future of Defence: Who Wins and Who Loses

    1. DCDavid Cahn58:29

       this change.

    2. HSHarry Stebbings58:30

       Dude, I'm- I'm throwing a curveball here, but I- I was told that you're the man who does defense at Sequoia. And I, you know, I, you know I say this with love, but I'm going in hard ball on this one. How would you respond to Sequoia were asleep at the wheel when it came to defense, not being in Helsing and Anduril, the two clear market leaders in the category?

    3. DCDavid Cahn58:52

       I would say, and I think this ties into our conversation so far, that AI- that defense is the next AI. I'm like, "That's how I started getting involved in AI." I think that defense is, if the transformer moment was sort of the starting gun in AI, uh, I think that, uh, the ChatGPT moment hasn't happened yet. So, I do think, look, there's no way around it, Sequoia was late to defense. Um, but I think Sequoia is working really hard to catch up, and that's part of business. You don't always get things right, but you keep trying, and I think we have that ethos and we have that humility. But what I would say is-

    4. HSHarry Stebbings59:20

       Why do you think defen- why do you think defense is the next AI? Sorry.

    5. DCDavid Cahn59:23

       So, I think that, you know, it's funny, because I started investing in AI, as we were talking about, a year after the transformer paper in- in 2018. And, um...You know, I think that it's sort of- re- defense reminds me, in some ways, of like a few years after the Transformer paper, which is to say, people who are really paying attention understand that AI- that defense is, is gonna change, and the Transformer moment was the U- was the Ukraine war. It was a very odd, you know ... Before that, you had to be a visionary, and, and to Palmer's credit, and, and Peter Thiel's credit, and people like this, like, they were visionaries. Before the Transformer paper, you're a visionary. Uh, and I ... You know, Ilya, Andrej Karpathy, these people are visionaries. After the Transformer paper, you're an early adopter, right? And I think our job as investors is to be early adopters, uh, for the most part, especially in the growth business, to be early adopters. And so, you see the, you see the change that happened in Ukraine, and, um, and I think it was very obvious that like, you know, warfare, you see these pictures of these tanks, you know, and these like long chains of tanks from Russia, and it's like, wow, like defense technology is 50 years old. And technology has moved so much in 50 years, and yet like, the way that we do war just hasn't changed, and that's because, um, you know, we've been in this period of, of golden era for the, for the world of, of dramatic peace and prosperity and all this stuff. And so, anyways, I think that the Transformer paper moment was (laughs) the, um, was the Ukraine war, and then I think the ChatGPT moment hasn't happened yet. And so I think their defense is actually, you know, it is under-hyped (laughs) in some ways, or like underestimated in some ways, and that's why I started getting in- interested in defense, uh, two years ago.

    6. HSHarry Stebbings1:00:48

       When you look forward at a world of AI, you assume that everyone will be improved with AI, will use it hundreds of times a day, and it'll be a part of everything that we do and think and, and say in many respects. Taking that view on defense then assumes this continuing conflict, uh, increases, not even decreases from where we are today. That would go against human cycles. There are periods of intense conflict, periods of not. Suggesting that defense like AI would suggest that that is the case. How do you, how do you feel about that?

    7. DCDavid Cahn1:01:20

       Yeah. So I think ... By the way, I'll, I'll share a little bit about how I got interested in defense, and, and, um-

    8. HSHarry Stebbings1:01:25

       Mm-hmm.

    9. DCDavid Cahn1:01:25

       ... you and I (laughs) know each other now, so like before I got in AI, I was reading all this stuff and I'm trying to learn from people, and I, I think my sort of investing style is like, you spend two years learning about the thing, and then you kind of start investing in the thing. And so I, I sort of take my time to sort of build a foundation, and my foundation in defense is like reading Napoleon and Churchill and like all of the, you know, the history of war, you know, the history of wars, the history of defense, like geopolitics, really like getting educated, and I probably spent two years really educating myself, and meeting founders, and you learn a lot from founders, on this space before I got involved. And the thing that I learned, and I think the thing that a lot of people who are deeper in this space than I am already understand is, uh, deterrence is, is the, is the first thing. Um, you know, uh, you, you only go to war because you have to. The whole point of defense is to prevent wars. And geopolitics is a real thing, and there's like real co- competition between nation states, and that's always, that, that will continue. And so, as the world gets reshaped, and we are living through a reshaping of the world order, I think that's something that a lot of people have seen, have written about. There's a lot of variables about that that we can unpack. Uh, I think Ray Dalio's ch- Principles of the Changing World Order is a really good book on this topic. So world order is, is sort of fundamentally changing, and that leads to this interesting opportunity where, um, we have to sort of catch up. There's a 50 years of catchup that has to happen. That's where I see the current defense moment. And this is why I say we're (laughs) you know, e- y- two years after the Transformer paper, we're not even at the ChatGPT moment yet, is, we're like one percent there on catching up. Like we're actually so, so early (laughs) in this defense cycle, because, you know, n- now we have a few dozen companies, maybe a hundred companies that have sort of new innovations. They're not integrated into the fore-structure meaningfully yet. There's so much more that has to happen, and I think that we have our, you know, the clear market leader now in the United States with Anduril, and I think there's more companies, uh, internationally that are gonna do really well as well. And, but I think that we're like, you know, we've, we've sort of (laughs) crossed the chasm of like, this is a thing that matters. We've crossed the chasm of, the government knows this matters. We've crossed the chasm of, you talk to people in Washington DC, they now understand Palantir and Anduril, they know those businesses. But in terms of the fore-structure changing, in terms of the way that we actually protect ourselves changing, in terms of US deterrence changing, I don't think it's changed that meaningfully. And I think after the ChatGPT moment, what's gonna happen is that, you know, pre-ChatGPT, if you were paying attention you noticed. After ChatGBT, everyone knew this was important, every American, every single person, and I do think we're gonna get to a place in defense where everybody knows that this is really, really important and that we need these companies to succeed.

    10. HSHarry Stebbings1:03:53

        Do you not worry about the concentration of buyers in that world? Again, when you compare it to defense, you have every business in the world, or every consumer in the world. What I really don't like with defense is actually what Bryan Singelman told me about what makes Anduril so special, which is a complementary skill set of the founding team, you know, whether it be GTM into, uh, like defense and government, whether it be product, whether it be intense ops with, um, you know, their CEO, Brian Schimpf. Um, and I, I just don't like the concentration of buyers and the selling to governments and the lack of incentive for them. Do you not worry about that?

Episode duration: 1:13:28