Skip to content
Uncapped with Jack AltmanUncapped with Jack Altman

David George on Growth Investing, AI, and Why the Power Law Is Stronger Than Ever | Ep. 58

David George is a General Partner at Andreessen Horowitz, where he leads the firm's growth fund. Before a16z, David was a General Partner at General Atlantic. He has been involved in investments including Databricks, Ramp, Harvey, and Figma. We discussed David's framework for thinking about the AI buildout: why the thesis that "it's all going to work" is the right one, and why the answer to almost every question in AI right now is "and" rather than "or." We got into where we actually are in the diffusion of AI into the enterprise, why there are 1.5 billion knowledge workers that AI has barely touched, and what would have to be true for the buildout not to continue. David shared his framework for thinking about product cycles vs. capital cycles, why right now is a 9 or 10 out of 10 on the product side, and why autonomous driving and robotics are massively underappreciated. We also talked about why vibes and narrative matter more than ever for founders, why he never shorts a messianic founder or a product people love, and how he thinks about the scale and ambition of Andreessen Horowitz as a firm. Timestamps: (0:00) Intro (0:23) The answer to every AI question is "and" (1:27) Where we are in the AI buildout (4:55) What would stop the token buildout (7:16) Sellers vs. buyers of tokens (10:39) Frontier labs vs. open source (13:18) Application investing and the coding blast radius (14:01) Harvey and legal as a category (20:46) Consumer AI and where we are (22:14) From reactive to proactive AI (25:52) AI, autonomy, robotics, bio health (27:49) Autonomous driving and why it's underappreciated (30:46) Robotics and what comes next (33:25) Benchmark's new growth fund (34:11) Why growth investing is compelling now (34:53) Half of private market returns happen at growth (37:57) Product cycle vs. capital cycle (41:54) The importance of narrative (43:14) Why vibes matter (46:20) Never short a messianic founder (49:39) How big can Andreessen Horowitz get Links: https://x.com/DavidGeorge83 https://x.com/jaltma https://uncappedpod.com/ friends@uncappedpod.com

David GeorgeguestJack Altmanhost
Oct 1, 202653mWatch on YouTube ↗

EVERY SPOKEN WORD

  1. 0:00 – 0:23

    Intro

    1. DG

      If the premise of your question is, is this gonna be successful or that thing gonna be successful? The answer in AI is probably and.

    2. JA

      Yeah.

    3. DG

      Like, it's just the most exciting time, you know, to ever be an investor and in the technology markets. Mm.

    4. JA

      And keeping up with it is gonna be hard.

    5. DG

      Keeping up with it is gonna be hard. [upbeat music]

    6. JA

      All right, David, I'm really excited to do this. I've been, I've been really looking forward to this. So thanks for, thanks for coming by.

    7. DG

      Yeah. Great to hang with

  2. 0:23 – 1:27

    The answer to every AI question is "and"

    1. DG

      you.

    2. JA

      So th- uh, one of the topics that's been on my mind a lot, um, that I think you've got really good thoughts on is something that my partner Eric Vischer said recently on, um, the Invest Like The Best podcast. And it was basically, he was like, "It's all gonna work." And it was in the context of, you know, people question, is it gonna be open source or frontier? And he's like, "The open source is gonna work. The frontier is gonna work." Is it gonna be Nvidia or is it gonna be new chips companies? And he's like, "There's gonna be both." And you can kinda like go on and on through, you know, the stack in AI. And Eric's point was basically, like, people are asking the wrong question of just like, is it this or it's that? He's like, "It's both." And I'm curious how you think about this at Andreessen and sort of in your seat.

    3. DG

      Yeah. Of course. So I am very, very closely aligned with Eric on this point that it's all gonna work. I, I actually like the way that he framed it up. I framed it up slightly differently, which is just like the answer to all this is and. And so, like, if you're answering-- if your, if your question, if the premise of your question is, you know, is this gonna be successful or that thing gonna be successful? The answer in AI is probably and. Um, I think it's helpful to start

  3. 1:27 – 4:55

    Where we are in the AI buildout

    1. DG

      just like, where are we? So, um, right now, from an infrastructure buildout standpoint, we just surpassed the railroads.

    2. JA

      Right.

    3. DG

      As, as a percentage of GDP, which has been well covered.

    4. JA

      It's like 3% now or something like that.

    5. DG

      Yeah. Exactly.

    6. JA

      Yeah.

    7. DG

      And, um, you know, look, it's probably gonna get harder and harder to build data centers, which we could talk about later. But I think that's gonna 10X. So right now, like what's happening is basically-

    8. JA

      As a percentage of GDP, you think?

    9. DG

      I think it will t- no, I think hopefully that will drive GDP growth.

    10. JA

      In dollars.

    11. DG

      And this is gonna take many years. So it won't 10X in one year.

    12. JA

      Yeah, yeah.

    13. DG

      So it will be, you know, whatever, 30%.

    14. JA

      But we'll be spending many trillions.

    15. DG

      But we will spend many trillions, right? So like all of this success, which is the greatest, fastest success we've ever seen in business, which is basically OpenAI, Anthropic together, and then if you add in SpaceX AI. Now with Cursor, they're call it like 120 billion of revenue, and they add more revenue per month than all the hyperscalers, except for Amazon.

    16. JA

      That's crazy.

    17. DG

      Now, all of that is on the back of just actually a small amount of people paying for stuff. So it's like 30 million coders in the world who are the vast majority of the revenue. There's 1.5 billion knowledge workers.

    18. JA

      Yeah.

    19. DG

      But like basically all of this, which is like adding more-

    20. JA

      Yeah

    21. DG

      ... more revenue than the best businesses that have ever been created-

    22. JA

      Yeah, yeah, yeah

    23. DG

      ... is on the back of a small group of people.

    24. JA

      That's actually a really interesting point, that it's over a billion users of these things, and we think about it that way. But it's actually, you're saying it's 30 million are spending basically all the money.

    25. DG

      It's 30 million. So there's a billion consumer users. And by the way, we should talk about consumer because we're like nowhere in consumer right now.

    26. JA

      Yeah.

    27. DG

      Even though there's a billion users. But in the enterprise, basically the, the vast majority of the revenue is coming from just a small amount of coders.

    28. JA

      Yeah.

    29. DG

      And this is like in power law-

    30. JA

      How many are within that 30 million? I'm sure it's a power law.

  4. 4:55 – 7:16

    What would stop the token buildout

    1. DG

      it generally gets used.

    2. JA

      So I... So ju- just to sort of like check the premises along the way, and obviously, as you know, I fully agree with what you're saying.

    3. DG

      [chuckles]

    4. JA

      But, um, just to check it along the way, what would have to happen for it not to be the case that as gigawatts come online, all the tokens keep getting consumed? Is there any world where... Is there anything that could update our priors, you know, that four gigawatts becomes 20 and they don't all get used at the same or even higher prices?

    5. DG

      I think there's two forms where this could take hold. One is, um, like what happened with coding is anomalous relative to everything else in B2B or knowledge work. And because of the difficulty of injecting commerce from knowledge work from companies, like the rest of white collar work just doesn't really like have major impact. Like I think that is extremely unlikely. Um, I think it's sort of like a data collection problem-

    6. JA

      Yeah, it's too hard to do

    7. DG

      ... and like an application of the models problem. The other would be, you know, we do have teams that we've backed that are working on like algorithmic breakthroughs, right? And so, you know, perhaps you would get an algorithmic breakthrough where, you know, rather than having to consume-

    8. JA

      Oh, just efficiency goes way up

    9. DG

      ... you know, massive... Yeah, like efficiency goes way up-

    10. JA

      Yeah

    11. DG

      ... both in training and in, and in inference.

    12. JA

      Yeah.

    13. DG

      Um, I think that's unlikely. I think it's possible we could get breakthroughs in training. Like I don't know, you've probably seen the stats on what it took you and I to like build our knowledge base.

    14. JA

      Yeah.

    15. DG

      Like it's like, you know, whatever, two or three orders of magnitude less than what the models actually train on. Um, so there seems to be an opportunity there. You still will then have inference, which is an absolutely massive market that's gonna be much bigger than training anyways.

    16. JA

      Yeah.

    17. DG

      And so, you know, yeah, we could get efficiency gains in inference, but I think what we've learned so far is the more you reason at the model-

    18. JA

      Yeah

    19. DG

      ... like w- at the time of inference, the better the results get.

    20. JA

      Yeah.

    21. DG

      And so I think that'll continue to hold, um, in knowledge work broadly. So I, I struggle to find a way-

    22. JA

      Yeah

    23. DG

      ... for this, like, not to work. Now, if we went to, you know, from 5 to 100 in a quick amount of time-

    24. JA

      Sure

    25. DG

      ... and the products aren't built to actually, like, take that on, you know. Sure. But, like, we don't-- we're not gonna have that problem, because it's gonna be really hard to get this stuff online.

    26. JA

      I'm curious what you think about this. I mean, the, uh... So I also can't easily see it, but the best vector that I've heard was a conversation I had with a friend where we were basically saying, you know, all this stuff come, has come online. And so at first, you know, a few years ago, there was, like, the big build-out, and you saw, like, you know, s- we're putting hundreds of billions

  5. 7:16 – 10:39

    Sellers vs. buyers of tokens

    1. JA

      in at the time. And then you get, you know, the models get a little bit better, and you start deploying them, and you kinda like keep moving through the pipeline. And now you see revenues growing like crazy on the back of AI. But the point in this conversation was most of the revenues that are, that we're talking about are sellers of tokens, not buyers of tokens. And so the next thing we kind of need to see is data around the buyers of tokens actually getting these huge efficiency gains. In other words, you know, a hyperscaler or-

    2. DG

      Yeah, like what's the ROI at the en- enterprise level?

    3. JA

      Yes.

    4. DG

      Like, you know, users, like, w- of writing all this more code.

    5. JA

      Yes.

    6. DG

      Like, is your bus- is, is your business getting better by writing all this more code?

    7. JA

      Yeah. Or said another way, it would be a bad outcome if the sellers of tokens were making lots of revenue and the buyers of tokens were not making lots of revenue-

    8. DG

      Yeah. Yeah, yeah, yeah

    9. JA

      ... but they were just buying lots of tokens.

    10. DG

      Yeah. Yeah, yeah. Look, I, I, I'm very optimistic about this point. Like, I think that whole argument is predicated on the fact that there won't actually be tangible, tangible productive use of, like, writing more code-

    11. JA

      Yeah

    12. DG

      ... far more efficiently that is better.

    13. JA

      Right.

    14. DG

      Like, I just think that's, like, fundamentally wrong. Um.

    15. JA

      Yeah, and I haven't, I haven't looked. I know like a year ago, there was a study, or it might even been a little more, but it's some- something like a year ago. There was a study that, like, in AI coding, it was, you know, actually you got 20% less effective by using AI code. But this was like before-

    16. DG

      This was before the big four five-

    17. JA

      Yeah

    18. DG

      ... AI breakthrough, right?

    19. JA

      And my, my guess is if you reran that study, you would not see that anymore.

    20. DG

      Yeah, and like the best coders in the world have, like, fully flipped over on this.

    21. JA

      Yeah.

    22. DG

      And, um, you know, if you talk to, like, the CEOs of the cutting edge companies, I think they would tell you that they feel much more productive.

    23. JA

      Yeah.

    24. DG

      Like, I, you know, spending time with Stripe, like, they feel like there's very high returns-

    25. JA

      Yeah

    26. DG

      ... to them being able to be a lot more efficient in writing code. Um, you know, like, is, like, Procter & Gamble getting big efficiency gains from what they're spending on AI today? Like, probably not yet.

    27. JA

      Yeah.

    28. DG

      But I think they will.

    29. JA

      Yeah.

    30. DG

      They probably need a lot of hand-holding. That's probably the answer.

  6. 10:39 – 13:18

    Frontier labs vs. open source

    1. DG

      done with this stuff. And by the way, we're just talking about enterprise right now.

    2. JA

      Right.

    3. DG

      Like, we talk, we should talk about consumer, but just in enterprise. So, you know, if you look at the Frontier labs, like today, you know, the Frontier labs are like 95% plus of dollars.

    4. JA

      Yeah.

    5. DG

      Like the 125 billion or so across the three of them. Um, you know, it's like they're consuming all of the revenue in the market, basically. Everything else is surrounding error. They have raised more than, like, the entire downstream ecosystem. Like just OpenAI and Anthropic have raised like $350 billion. So, like they, they do have scale.

    6. JA

      Wow.

    7. DG

      Like they have massive scale benefits.

    8. JA

      Yeah, yeah.

    9. DG

      Um, and right now, what's revealed in the preferences of users is, um, users are opting to use Frontier models, and I think there's a bunch of reasons why that's the case. Um, right now people are willing to pay, you know, 10X more for Frontier, for Frontier tokens. Um, you know, I think in coding that's, that's definitely gonna persist for some period of time. Um, you know, there's, there's reasons why you could see a case that people will like cost optimize and-

    10. JA

      Yeah

    11. DG

      ... and use like below the Frontier over time. Um, but right now like-

    12. JA

      Yeah, but too early to have to assume that the... Yeah

    13. DG

      ... the revealed, the reve- revealed preference is like everyone-

    14. JA

      Yeah

    15. DG

      ... everyone who is a heavy user of this stuff wants to use the Frontier. I think there's-

    16. JA

      Well, and also it seems to me like if that happens, the Frontier should be at even more valuable tasks that are worth even more. You know, like if coding-

    17. DG

      Even more so, like, yeah, curing cancer. Yeah, like if-

    18. JA

      If coding was worth X, like what's curing cancer worth?

    19. DG

      Curing cancer is worth, you know-

    20. JA

      More than X

    21. DG

      ... some, some astronomical. Yes, of course. And so right now, like I think there's a couple reasons why people are using the Frontier stuff. Like one is just there's so much user or consumer surplus to having this work done for you that you're willing to pay s- so much more, like you're willing to pay-

    22. JA

      Yes

    23. DG

      ... the, the incremental amount for that Frontier.

    24. JA

      Yeah.

    25. DG

      Does that make sense?

    26. JA

      Yes.

    27. DG

      Um, secondly, like they've done a pretty good job with first party products. So, you know- People really like using Codex and Claude Code and Cursor and, you know, I think increasingly we'll use stuff like GroqBot. Like, it's gonna be really valuable to use the first party stuff. There's a bunch of reasons why that's the case. They've done a good job with product. I also think they have very tightly coupled their harnesses with the models.

    28. JA

      Yeah.

    29. DG

      So that, um, the experience you have with those products is actually very, very good. And so subbing out some other model where you don't have the harness and you don't have the product and don't have the interface, like, no one's doing that today.

    30. JA

      I'm curious your thoughts on what you... H- how do you... And obviously it's very hard to know, but how, how do you sort of, um, mentally picture what their incentive should be on making more first party products? And I'm asking this in relation to as you think about investing in applications at the growth stage and you think like what's gonna be worth it for them to build a first party app in and will that matter or again does it all win? Like how do you think about their incentives and how if at all it relates to

  7. 13:18 – 14:01

    Application investing and the coding blast radius

    1. JA

      your application investing?

    2. DG

      Yeah. So we've done a lot of application investing. So again, this is like in the theme of and.

    3. JA

      Mm-hmm.

    4. DG

      Right? Like, um, we th- we think, we think a lot of things are gonna work. Um, on the application side, my framework is the big labs are going to be very focused on first party products that are obviously in coding and within the blast radius of coding. And so you can interpret the blast radius of coding however you'd like.

    5. JA

      Yeah. Yeah. Yeah, yeah.

    6. DG

      But it's, it's probably some things around it. Um, you know, uh, that's, that's one piece. And then secondly, they're gonna be focused on first party products that appeal horizontally to all 1.5 billion knowledge workers. So think like Microsoft Office and Google Apps, and they would like to have those as first party products. We know from their work that they're doing, like the,

  8. 14:01 – 20:46

    Harvey and legal as a category

    1. DG

      the work that they do with data vendors, like they are very focused in those areas, right? So think like, you know, all the Microsoft Office products. Um, and so, you know, outside of that, there's a question of like what number priority is it for them? So, you know, we're large investors in Harvey.

    2. JA

      Yeah.

    3. DG

      And so, you know, Harvey, I feel like, you know, legal is, uh, I don't know, it's very interesting to people on X. Um, but um, but it's a market that's like absolutely in takeoff.

    4. JA

      Yeah.

    5. DG

      Right? So if you take an optimistic view of it, there's many differences which we could talk about, but it's probably like 12 months behind coding.

    6. JA

      Yeah.

    7. DG

      Just in terms of like diffusion into the workforce.

    8. JA

      Yeah.

    9. DG

      And it's now gotten to the point where end clients are demanding that their law firms use Harvey.

    10. JA

      Yeah.

    11. DG

      So like we care about it for product and we care about it for cost. Um, and so you must use it. So it used to be this thing where people would talk about like hallucinations and then now it's just like, no, everyone has it. And so, um, you know, it's probably like somewhere between number six through 15 on the priority list.

    12. JA

      Yeah. By the way, that, that one, it, it really is funny how much it gets people on X and you know, we know this too obviously from-

    13. DG

      Yeah, from Agora. Yeah

    14. JA

      ... yeah. And, and it's like, um, I think people, A, were surprised that these companies can grow this fast in legal, which you would not have thought would be an early adopter and then you have a, you know, then you've got a lot of doubt, you know, 'cause they're like, "Oh, well wasn't it hallucinating and stuff?" It's like, yeah, like 18 months ago.

    15. DG

      Yeah, 18 months ago.

    16. JA

      But like it's-

    17. DG

      And like they made a smart bet, which is like what happens when models get 100X better?

    18. JA

      Yeah.

    19. DG

      And like, and you know-

    20. JA

      And it's gonna work. Yeah

    21. DG

      ... and it's gonna work.

    22. JA

      Yeah.

    23. DG

      Like and so the big break-

    24. JA

      And legal's actually just like a perfect thing too because it's language but it's like kinda mathematical. Like it's extremely logical.

    25. DG

      Yes, exactly. Like they're...

    26. JA

      Yeah.

    27. DG

      Yeah, it's extremely logical. It's extremely well documented.

    28. JA

      Yeah.

    29. DG

      Like in terms of case law.

    30. JA

      Yeah.

  9. 20:46 – 22:14

    Consumer AI and where we are

    1. JA

      of consumer and how do you see it?

    2. DG

      Yeah. I mean, there's a billion people using it, right? Um, and there's more than a billion people using it, you know?

    3. JA

      Obsessively.

    4. DG

      Obsessive- obsessively. Um, but like the instantiation of the product and the use case is like very basic, right? We're in like skeuomorphic mode for the vast majority of consumers, which is they are doing what they used to do on a search engine inside ChatGPT.

    5. JA

      Yeah.

    6. DG

      And like that's the extent of the use case of the vast majority of the-

    7. JA

      Except for the kids. The kids use it better than that.

    8. DG

      My... Yeah. My son, uh, I, I mean, total aside, a 10-year-old is like writing full-on rap songs-

    9. JA

      Yeah

    10. DG

      ... and creating music.

    11. JA

      Yeah.

    12. DG

      And, you know-

    13. JA

      No, it's not an aside. This is the future.

    14. DG

      This is the future.

    15. JA

      But this is-

    16. DG

      And it's like incredible

    17. JA

      ... it tells you the difference where it's like there's a, there's a, a, a large cohort of people who, you know, are gonna probably always Google, you know, use this as a Google replacement.

    18. DG

      Yeah.

    19. JA

      But that's not how the kids are gonna use it.

    20. DG

      The kids are gonna be, the kids are gonna be totally native.

    21. JA

      Yes.

    22. DG

      Like, and, and, you know, you can see it with the kids now. Like they didn't, they, they never were like trained, my kids were never trained on like search engines-

    23. JA

      Yeah

    24. DG

      ... and now they just, it's a whole new thing where they get it to, you know, to give them answers to anything and give them, you know, work product of anything. Y- the big shift is gonna be like when it goes from being reactive to proactive and, you know, we're like just scratching the surface on that.

    25. JA

      Yeah.

    26. DG

      Um, it's gonna be multimodal, so it's not gonna be just like sitting doing this. Like the voice stuff actually is very, very good now. Um, you know, there was a bit of a head fake because the OpenAI voice stuff was not very good for a period of time, and it had to

  10. 22:14 – 25:52

    From reactive to proactive AI

    1. DG

      rely on a dumber model. And now it doesn't and it can handle interruptions and things like that. So like a natural human interaction form now exists.

    2. JA

      Yeah.

    3. DG

      Um, and capabilities for doing proactive work on your behalf and like action taking is like, we're just now there, right?

    4. JA

      Yeah.

    5. DG

      Like OpenClaw was the first moment. I think GrokBock is a, GrokBot is a moment now. Um, and I think that'll be kinda the future of consumer use. It's so funny to me that no one talks about consumer right now.

    6. JA

      Yeah.

    7. DG

      Like everyone is obsessed with coding, rightfully so-

    8. JA

      Yeah

    9. DG

      ... given what's happened with the revenue ramps of the companies. But if like historical technology markets are any guide-

    10. JA

      Yes

    11. DG

      ... like the big prize is gonna be the killer consumer products 'cause you can, you can address billions of people with them. So I think those are gonna get a lot better. I'm very optimistic. Um, and those are gonna come with like the world's best business models too. There's gonna be like a lot of value generated. There's gonna be a ton of consumer surplus.

    12. JA

      Like what would that look like?

    13. DG

      I think it'll be advertising or some greater form of advertising. Like i- i- if you tried to describe... The thing I always say is like if you tried to describe the form factor of the feed to people, like during classifieds in newspapers, like it would make no sense.

    14. JA

      Yeah.

    15. DG

      So we have yet to figure out like what the native consumption model is gonna be and then what the native ad format is gonna be that follows that.

    16. JA

      Yeah.

    17. DG

      Um-

    18. JA

      Do you think it's ads and subscriptions?

    19. DG

      Yeah, definitely.

    20. JA

      Yeah.

    21. DG

      Yeah. I think it's both.

    22. JA

      Like you think that it'll be a situation where hundreds of millions of people might be willing to pay today surprising amounts for this thing?

    23. DG

      Yes, absolutely.

    24. JA

      Yeah.

    25. DG

      And again, I think-

    26. JA

      And that's when it's like a personal assistant for everybody.

    27. DG

      That's when a person, it's a personal assistant for everybody.

    28. JA

      Yeah.

    29. DG

      And so like I, again, I think my framework is like 10X more value will just accrete to the surplus of the users.

    30. JA

      Oh, wow.

  11. 25:52 – 27:49

    AI, autonomy, robotics, bio health

    1. JA

      the autonomy curve with that?

    2. DG

      Yeah.

    3. JA

      Yeah.

    4. DG

      It's so funny to think about these other major technological tailwinds that are happening because like AI just, you know, like supersedes everything-

    5. JA

      Yeah

    6. DG

      ... in our minds right now.

    7. JA

      Totally.

    8. DG

      Um, like I always find it helpful to just take a step back and just say like, okay, what are like the wa- like product cycles drive our business. And so like what are the big waves that we're catching right now? Like autonomous driving is one. But the previous generation of these was like mobile phones plus social, e-commerce, SaaS, cloud, kind of all coinciding.

    9. JA

      Yep.

    10. DG

      And that produced like $25 trillion of market cap.

    11. JA

      That's a lot, yeah.

    12. DG

      And like it's like a ton of, that's like a lot, right?

    13. JA

      Yeah.

    14. DG

      Um, and you know, like what do we have staring at us in the face right now? Like we've got AI, we've got autonomy, we've got robotics, we've got AI's application to health.

    15. JA

      Yeah, I was gonna say bio.

    16. DG

      Like we have the American dynamism category, which is like-

    17. JA

      Yeah

    18. DG

      ... you know, modernization of defense.

    19. JA

      Yep.

    20. DG

      Like the combination of all of those, I'm very confident-

    21. JA

      Gotta be more than 25

    22. DG

      ... it's way more than 25.

    23. JA

      Yeah.

    24. DG

      Like i- it's just the most exciting time, you know, to ever be an investor and in the technology markets. Autonomy specifically, I, I think it's just very underappreciated. Like the auto industry and people driving cars is one of the biggest industries in the world, obviously. Um, right now, okay, so if you rewind all the way to like taxi driving, right? Um, that was like less than 0.1% of the overall miles traveled in the US.

    25. JA

      Yeah.

    26. DG

      And there were a bunch of reasons why that was the case. Then the rideshare companies came along and, you know, I was an investor in Uber, you know, obviously y- you guys were too. Um, the big thing that happened obviously was like a better business model, but it was, it was really like the 10X'ing of the market size. So like, you know, taxis were 100 million bucks in San Francisco, and then they became a billion dollars like three years into Uber and Lyft. Um, and so it 10X'd the market size because you delivered a much better product. Autonomous driving, we're gonna look back 10 years from now and say like, "Oh, there was another obvious 10X'ing that

  12. 27:49 – 30:46

    Autonomous driving and why it's underappreciated

    1. DG

      would happen there." Um-

    2. JA

      At least.

    3. DG

      At l- m- at least just on the ride hail side. So if you have a product that is 14 times safer than a human driver-

    4. JA

      Then it's barbaric to drive.

    5. DG

      It's actually, so like this is my view now on the cities-

    6. JA

      Like don't you think like if you're like, how many people die a year from a human driver? If it's safer, isn't that-

    7. DG

      You know-

    8. JA

      'Cause isn't the data like pretty abundant now?

    9. DG

      It's 14, it's, it's millions of dri- millions of miles traveled in Waymos, and they're like between 10 and 14 times safer than a human driver.

    10. JA

      Yeah.

    11. DG

      So like it is barbaric to not actually allow for these things to diffuse into your economy. So, um, I think like history will look back on people and judge them very poorly if they don't let it happen.

    12. JA

      Yeah.

    13. DG

      Um, but for the consumer side, just start with that. If you can drive in something that's 10 to 14 times safer, you would pay a massive premium.

    14. JA

      Yeah.

    15. DG

      I think there's a very clear case for, um, how you actually get the cost below the cost of Uber and Lyft today. Um, so right now, owning a consumer car equates to like 80 cents per mile, like it fully loaded, depreciation, insurance, gas, all that stuff. Um, like that's your like cost of an incremental mile. Um, riding in an Uber or Lyft is probably like two bucks and change per mile. Um, this market is like massively elastic. So like if you can get, you know-

    16. JA

      The cost down

    17. DG

      ... the cost of a Waymo or a Tesla or whatever, you know, ride hail service-

    18. JA

      Everything

    19. DG

      ... which I think there'll be others.

    20. JA

      Yeah.

    21. DG

      Um, down below that, like there's going to be an explosion of demand. So it's 14 times safer. It's only gonna get better, by the way. It's gonna get safer. It's gonna be 20 times safer soon. Um, and you know, it's gonna be less than the cost. Like you will see a 10X plus.

    22. JA

      Yeah.

    23. DG

      Then you talk about personal car ownership. Uh, and so there's 17 million cars sold a year in the US new. Um, and by the way, there's like 250 million cars in the US. Um, just in the US, like how much would you pay for full autonomous features for a, a passenger car that you owned?

    24. JA

      A lot. It's like double.

    25. DG

      10,000. Uh, yeah, like minimum $10,000.

    26. JA

      Yeah. Yeah. Yeah.

    27. DG

      Right? Um, and so-

    28. JA

      I mean, once you're used to it, it is extr- and I can't even-

    29. DG

      Yeah, you can't, can't not have it

    30. JA

      ... it's really hard to go back. Yeah.

  13. 30:46 – 33:25

    Robotics and what comes next

    1. DG

      huge markets, um, that we're super excited about.

    2. JA

      Yeah, and then, you know, related, there's robotics, which is gonna be enor- I mean, not, there's nothing quite going yet. I mean, but you see the glimmers of it now, and it's like you go into the labs that are working on it and, you know, like the startups that are doing it. Obviously, Tesla's very good at it. There's... But it's, it's clearly gonna work.

    3. DG

      It's clearly gonna work, yeah. And, like, it, it'll be a bigger market than language, right?

    4. JA

      Yeah.

    5. DG

      When it does work.

    6. JA

      Yeah.

    7. DG

      Um, because it will have, you know, B2B, you know, applications. It'll have consumer applications. Like, I think we're probably pretty far off-

    8. JA

      Yeah

    9. DG

      ... from having-

    10. JA

      It does feel like we're-

    11. DG

      ... ro- robots in our house.

    12. JA

      Yeah.

    13. DG

      Um, but you know, for-

    14. JA

      But for like-

    15. DG

      Like, defined use cas-

    16. JA

      ... you know, in a factory.

    17. DG

      Yeah, factory.

    18. JA

      Yeah, exactly. Yeah.

    19. DG

      Like, relatively defined use cases with, like, redundant tasks with safe environments.

    20. JA

      Yeah.

    21. DG

      Like, we see it.

    22. JA

      Yeah.

    23. DG

      Like, we invested in, um, you know, one of my partners led an investment in, um, uh, in mind robotics. And so, you know, s- Sarah did this deal, and it's the founder of Rivian, and, you know, he basically was like... First of all, he's an exceptional founder.

    24. JA

      Yeah.

    25. DG

      Um, and has produced a great company with great products. Um, but he's going to put robots on the factory floor, and so they'll do, you know, manufacturing assembly work on Rivians. It's relatively, you know, defined use cases with pretty high ROI today, and then it's an embedded customer. You know, you kinda have it, and you can not only get revenue from that, but you can also have the models learn from the work that you're doing.

    26. JA

      Yes.

    27. DG

      And so you have this feedback loop in the field that you can actually work on. Um, you know, there's other stuff that is kinda picks and shovels stuff that we're excited about, but, like, we will have a ChatGPT moment for robotics. I don't know when it's gonna be.

    28. JA

      Yeah.

    29. DG

      Like, I think it's within five years, hopefully.

    30. JA

      Yeah.

  14. 33:25 – 34:11

    Benchmark's new growth fund

    1. DG

      I think it's gonna be a massive productivity boom. It's gonna be awesome.

    2. JA

      So I wanna talk about capital and financing a little bit.

    3. DG

      Yeah.

    4. JA

      We, we at, uh, Benchmark just raised our first growth fund in history, and, you know, the, the baseline for that was something... Yeah.

    5. DG

      We- welcome to the dark o- w- welcome to the dark side.

    6. JA

      Yeah, exactly.

    7. DG

      [laughs]

    8. JA

      It's like, uh, you know, the, uh, the forbidden fruit. And, um, some- you know, but the, the thinking for us behind it was something along the lines of venture returns now happen at growth in a way that they didn't used to.

    9. DG

      Yeah.

    10. JA

      And, you know, this conversation we were, that we've just had really supports it because, you know, the outcomes are that big, and the things that run really run, and it's all gonna work, and all of these things. And that sorta led us to the thinking that, you know, the sort of 10 to 100x or 1000x, whatever, you know, the, the, the really big multiples can happen all along the way.

    11. DG

      Yeah.

    12. JA

      Um, which I know

  15. 34:11 – 34:53

    Why growth investing is compelling now

    1. JA

      you believe that obviously.

    2. DG

      Preaching to the, preaching to the choir. [laughs]

    3. JA

      Um, I wanna connect this though. So, you know, I, I, I would like to hear your thinking on that, and I would also like to hear... You know, we just talked about the product cycle and where we are, but there's also this related but not exactly the same thing of the capital cycle.

    4. DG

      Yeah.

    5. JA

      And the capital following the products and investing behind them and all of those things. So, uh, I would just be curious to hear your take on where we're at from a capital perspective, from a growth investing perspective.

    6. DG

      Yeah, of course. I, so, um, you know, as it, as it relates to, like, why go do growth and, like, is it a compelling category, obviously, you, you have, you have seen the light. Um, but, um-

    7. JA

      The dark light

    8. DG

      ... the, the dark, it's, it's great. Um, i- the market that we're in is, like, a $5 trillion market, so it's

  16. 34:53 – 37:57

    Half of private market returns happen at growth

    1. DG

      like, you know, whatever, it's like 20% of the S&P 500. So if you're not actually, you know, playing the full, you know, the full sort of gamut of that in the private markets, like, you're probably leaving a lot of money on the table. Like, we did this analysis a long time ago that basically showed, um, half of, half of private market returns get generated between the seed and the B, and then half of returns get generated from the C+. And so, you know, if you're not aggressively playing from the C+, you're-

    2. JA

      You miss half

    3. DG

      ... you're, you're definitely leaving off half. And by the way, that was during an era when companies went public earlier. So now that companies are staying private longer, if you could fast-forward five years, the ratio's gonna be even more extreme. It'd probably be 70/30 or something in favor of the late stuff. Um, you can also give venture outcomes because the power law is so crazy strong today, right? And so, you know, when you get it right, like, it can really go right. And we're in this dynamic market where the power law is more extreme than in the past, and I think it's a result of a few things. One, um, is just, you know, we're, we're early cycle, and so, you know, the, the early companies can establish themselves as leaders earlier. Um, secondly, this one in AI is unique because if you throw more money at it, it can just scale up-

    4. JA

      Yeah

    5. DG

      ... and get better, right? So, like, that is different than previous cycles, right? If you threw endless amounts of money at ServiceNow or Workday or Salesforce, like, during the SaaS era, they would get all messed up.

    6. JA

      Mm-hmm.

    7. DG

      Like, we saw this experiment with, like, the Vision Fund, right? Like, you throw money at these things, and the companies get screwed up, right? Or if it was a consumer company, they would spend it on Google and Facebook, and, like, that's not a good thing either.

    8. JA

      Was that the mechanism for why the Vision Fund was... Was it just, like, dumping too much money in companies that didn't need it, or was it hitting, you know, the cycle at too late of a time, or was it both?

    9. DG

      I think it was both. I think it was like- Well, first of all, like, I don't think it was a bad idea. Um, I think it was just, like, wrong time of the cycle, and then we were not in an environment where if you put more money into something, it got better-

    10. JA

      Yeah

    11. DG

      ... just automatically, like with, with no scale limitations. Like, that is what happens with AI.

    12. JA

      Yes.

    13. DG

      Like, it, it used to be, you know, if you put, you know, a $40 billion fundraise into Salesforce, like-

    14. JA

      The company's worse

    15. DG

      ... what are they gonna do? Like, go hire a bunch of sales reps and engineer? Like, it just wouldn't work.

    16. JA

      Yeah.

    17. DG

      Whereas if you put $40 billion towards training, you know, these models-

    18. JA

      Things get better. Yes

    19. DG

      ... you know? Like-

    20. JA

      Yeah

    21. DG

      ... they get much better.

    22. JA

      Yeah.

    23. DG

      And so that is very unique, uh, in the moment that we're in right now. You know, I, I think about the world as, like, product cycles and capital cycles and, um, I tell my team, like, the ideal state of the world would be you're in a great position on the product cycle, like scale of one to 10, 10 being good, one being not good. Like, in 2021, we were at, like, a one, which is like, we didn't realize it at the time.

    24. JA

      Yeah.

    25. DG

      Um, but that was-

    26. JA

      It was like-

    27. DG

      ... like, late SaaS-

    28. JA

      It had played out

    29. DG

      ... late cloud.

    30. JA

      Yeah.

  17. 37:57 – 41:54

    Product cycle vs. capital cycle

    1. DG

      it was, like, a eight out of 10.

    2. JA

      Yeah.

    3. DG

      Like, it was very good.

    4. JA

      Yeah.

    5. DG

      Um, so right now product cycle is, like, a nine or 10 out of 10.

    6. JA

      Yeah.

    7. DG

      For all those reasons I have laid out.

    8. JA

      Yes.

    9. DG

      Like AI, autonomy, American dynamism, robotics, bio health, like, all these things are gonna happen over the next 10 years. Um, capital cycle, you know, your ideal would be, like, a 10 on product cycle and, like, you know, a 10 on capital cycle, meaning valuations are really low.

    10. JA

      Yes.

    11. DG

      They almost never coincide-

    12. JA

      Yes

    13. DG

      ... because there's, like, excitement about the current-

    14. JA

      Yeah, it'd be like it was like the best part of the product cycle ever and no one else saw it but you.

    15. DG

      Like, 2023, like, early days of when AI started working-

    16. JA

      That was probably pretty good

    17. DG

      ... and the capital markets were a little more benign-

    18. JA

      Yeah

    19. DG

      ... and, like, some of the crazy stuff and run-ups hadn't happened yet, like, that was like, I don't know, like, a seven.

    20. JA

      Yeah.

    21. DG

      And you know, like, 2010, you know, you could buy, like, SaaS companies and B2B marketplace companies for, like, you know, whatever, four times revenue or something. Like, that was pretty good, too.

    22. JA

      Yeah.

    23. DG

      Um, right now capital cycle-wise, like, people see the excitement, and you can still find ways to do good investments, and the companies are growing so fast, and they're so dynamic, it's fine, but, like, we're probably at a six or something.

    24. JA

      Yeah.

    25. DG

      You know, like, it's, it's okay-

    26. JA

      Yes

    27. DG

      ... but not great. Um, but the thing that drives our business over 10 years, whether it's venture or growth, is the product cycles.

    28. JA

      Yeah.

    29. DG

      And that's, like, a nine or 10 out of 10 right now.

    30. JA

      Right. Yeah. It's sort of like, um, I think it's... I can't remember if this is, I heard this first from Thrive or somebody, but it was like, you know, one way to be contrarian is to believe a thing nobody else believes. The other way is to believe the same thing everybody else believes, but believe it 10 times more or something like that.

  18. 41:54 – 43:14

    The importance of narrative

    1. DG

      put in a financial model.

    2. JA

      Yeah.

    3. DG

      And, like, that is the thesis that we have.

    4. JA

      One of the things that I really look up to about Andreessen is I think you got the sort of importance of narrative right from the beginning. You know, like, I think about even when the firm was started, there was such an understanding that, you know, at the time it was, you know, media was very important, but there was just an understanding that stories matter and that, you know, the way you present matters, and being loud and in the conversation is really important. And I feel like obviously from, when from, what, 2009 or something?

    5. DG

      2009, yeah.

    6. JA

      You know, since then, obviously things have changed a lot and, you know, very, uh, I think in a very happy, uh, update, we've had sort of the lulification of everything where people are going direct and owning their own narrative, and you get more authenticity. You know, you guys have, uh, you know, a lot of, like, new media going and all these other things. But I think Andreessen has understood the importance of narrative

    7. DG

      Yeah

    8. JA

      And it seems to me that narrative is, like, louder than ever right now for companies. You know, it's like... And it's at all stages. It's like a company's launching, you see these beautiful videos, and companies at the later stages have their investors kind of, you know, creating stories for them in various ways, and they're owning media, and it's all just loud. But I think it's important, and I'm curious how you think about narrative now at this point in the cycle when we're all, you know, living on X

  19. 43:14 – 46:20

    Why vibes matter

    1. JA

      and we're living in, in the news and, you know. How do you think about this all?

    2. DG

      Yeah. Look, I, th- this is just, like, the vibes, right? Like, the vibes, the vibes matter tremendously.

    3. JA

      But they matter.

    4. DG

      They matter a lot.

    5. JA

      Yeah, right.

    6. DG

      I, I actually have a funny story from, um, a meeting that I was having with Ben in 2018. So this was, I don't know, three or four months before I joined, and we were kinda hashing out, you know, what is our strategy gonna be. And, uh, and so, you know, I had all these ideas of, like, these are the new teams we need to build and this is what the investment team looks like, and he's like, "Well, what's your marketing strategy?" And I was like, "We don't really need a marketing strategy. It's not really a thing in growth." And he just, like, looked at me and said, like, "That's the dumbest fucking idea I've ever heard." Uh, and like, venture firms didn't market themselves, and then we did and it worked really well, so, like, why would you not do it here? Um, and so I came along. You know, I came from a place that was a lot more conservative and, uh, and so I've, I've sort of seen the light. Um, vibes matter. Let's just talk about why the vibes matter. So the vibes matter because, um, of fundraising, of your valuation, um, for retaining your employees and for hiring new employees-

    7. JA

      Yeah

    8. DG

      ... among other reasons.

    9. JA

      Yeah.

    10. DG

      And, and, and, like, in some markets, the vibes matter for your customers too. But, like, at a minimum, it's those things. Um, if you look at, um... And, and this is basic, but, like, why does a higher valuation matter? Like, a higher valuation means, you know, less dilution, more ability to raise capital, um, and, um, you know, you can use that as a weapon if things go sideways.

    11. JA

      And in a world where capital actually makes your company better-

    12. DG

      Yeah

    13. JA

      ... it's worth more.

    14. DG

      Exactly. It's worth more. Um-

    15. JA

      It was worth less when it was not, you know, the thing.

    16. DG

      Yeah. And, like, now every investor and every employee and potential employee is, like, living on X-

    17. JA

      Yeah

    18. DG

      ... and, like, consuming the vibes. So if you look in the public markets, the companies that have done a great job of this, like, there's a direct line to their valuation and their life gets a lot easier with a higher valuation, right? So, like, famously right now it's, it's Palantir and Karp, and, um-

    19. JA

      Yeah

    20. DG

      ... you know, he's got, like, a cult following. We-

    21. JA

      You know what's amazing to me on that one, is that he... I mean, Palantir had a huge multiple before the recen- like, now it's... I think the growth has inflected recently, right?

    22. DG

      Yeah, it has. Yeah, it's massively accelerated.

    23. JA

      Massively accelerated.

    24. DG

      Yeah.

    25. JA

      But it had the premium before it had that.

    26. DG

      Yes.

    27. JA

      And there's something interesting there.

    28. DG

      Yeah. And, like, by the way, to tell you how much it's vibes, like, no one knows what ontology means.

    29. JA

      No, I don't know what that is.

    30. DG

      Right? It's not even, like, a thing, right? And, you know, but, like, the, the big thing obviously that happened recently-

  20. 46:20 – 49:39

    Never short a messianic founder

    1. DG

      to raise more capital, uh, to have loyal following. So you can look at, like, the high retail ownership stocks and there's, like, a, a correlation if they're doing well-

    2. JA

      Yeah

    3. DG

      ... a correlation to, you know, to the valuation. Um, uh, this is funny. I heard from a hedge fund guy the, this theory on shorting stocks, and his theory on shorting stocks was, "I never short a messianic founder and I never short a product that people love." The inverse of that is if you have a messianic founder and there is a product that people love, you, you can actually, like, trade very high and have all these benefits, right? Um, right now in AI, obviously you said why the capital side matters and valuation side matters. It is incredible that the whole market is, like, hanging by a thread on the performance of, like, two-

    4. JA

      I know

    5. DG

      ... two, now kind of three private companies-

    6. JA

      Yeah

    7. DG

      ... um, on, like, a monthly basis. Like, there's-

    8. JA

      I mean, I, I was just thinking as you were talking that the narrative, what- however strong the narrative is now, it, the importance of it all is about to go, like, hyperbolic as these companies get public.

    9. DG

      Yeah. And, um, you know-

    10. JA

      But you can even see it. You know, there's this funny bit I, I mentioned to you of, you know, you can see investors and, you know, people who are on the opposite side kind of jockeying about, like, trying to set expectations around these companies.

    11. DG

      Oh, yeah, like Anthropic and OpenAI's revenue and all that stuff. Yeah.

    12. JA

      Yeah. And it's like, but, you know, and it doesn't all... It's not like the insiders are all pumping, you know, crazy numbers. It's actually, like, expectation management versus, you know, and it's just, but it, the amount of calories going into controlling these stories is just gonna be wild.

    13. DG

      Yeah. And there are... It's not even just the CEO. So, like, I talked about how the asset class is the founder and I talked about Karp and I talked about Elon. Ideally, if you have a dynamic founder who can be the face of the company, it's like Palmer with Anduril-

    14. JA

      Yeah

    15. DG

      ... right? Like, that will benefit you in many ways.

    16. JA

      Yeah.

    17. DG

      Again, if you have a product that people love, that messianic founder thing is, like, a massive asset, and it will make people care at every, at every layer that matters, right? Capital side, employee side, customer side. Um, and if you don't have that, if you're, like, some no-name, like, SaaS CEO, it's, like, very hard to command people's attention and, like, make them care.

    18. JA

      Yeah.

    19. DG

      Um, it also matters for people down in the organization, right? So, like, right now I think the person who's doing the best job of this in the market is Tebow from OpenAI and, like, he's very direct, and, you know, not only is it feeding good vibes among the investor community and good vibes among employees, but it's driving a lot of behavior on the customer side too. And so I think when done right, um- You know, it can have massive benefits.

    20. JA

      Yeah.

    21. DG

      And it's not to be underestimated. It's something we talk to our founders about all the time now. It's like, "You gotta go direct. You gotta own the narrative. You're the face of the company. You need to tell a compelling story." And if you don't, yeah, some people will care, but a lot of people aren't gonna care.

    22. JA

      Yes. And someone else with a almost as good but not quite as good product can come along and tell the story better.

    23. DG

      Yeah, exactly.

    24. JA

      Yeah.

    25. DG

      Exactly.

    26. JA

      Yeah. My last question, and I, I think I asked, I had both Mark and Ben on, um, at different times, and they both were... You could just feel the ambition and the scale and sort of the mindset around all of it. And, um, you know, this shows up in a lot of ways and, you know, uh, you know, I was just joking to you before we sat down that, you know, Benchmark's pretty small, Andreessen's pretty big, you know, in terms of, like,

  21. 49:39 – 53:11

    How big can Andreessen Horowitz get

    1. JA

      its team and sort of all these things. Um, I guess my question is, how far, given that we're looking at not $25 trillion, but something bigger, like, how far can this all go? Like, how big can a venture firm, you know, get? What rate limiters, if any, are there on what you can do? Like, how do you think about, you know, what's left above, you know, the already, you know, large size you guys are at?

    2. DG

      Yeah. Look, we, we talk about this all the time 'cause there's p- there's pieces of our business that are scalable and there's pieces of our business that are not, right? And something that you said, which I think is super smart, is if you wanna go fast, you know, be small. If you wanna go far, you gotta be big, right? And so, um, there are things that we do that may cause, you know, more organizational overhead, but allow us, we think, to go further.

    3. JA

      Yeah.

    4. DG

      And, you know, pa- part of that is scaling our investment approach and scaling our team that does investing. And then a big part of that is the 650 employees that we have that help our companies. Um, so how big can it get? Yeah, $25 trillion of market cap creation. Um, tech is, call it, like, a third to a little higher percent of the overall market cap in the economy. Um, if you look at previous, um, eras, right, as a percentage of the overall market, different industries were actually much larger in the past. So, like, railroads and financial services and stuff. Like, they, they... At their peaks, they would be, like, 80% of the market. Um, I don't see a reason why tech will not continue to ascend as a percentage of the market cap. Like, right now, eight of the top 10 most valuable companies in the world are, you know-

    5. JA

      Yeah

    6. DG

      ... US venture, previously venture-backed technology companies in the US.

    7. JA

      Yeah, and they, and, and they all kinda look cheap.

    8. DG

      And they all look kinda cheap, and they're probably the best businesses ever created.

    9. JA

      Yeah.

    10. DG

      And then, and then by the way, like, we have OpenAI and Anthropic and SpaceX on their heels-

    11. JA

      Yeah

    12. DG

      ... that are, like-

    13. JA

      So you have more-

    14. DG

      ... growing faster.

    15. JA

      Yeah.

    16. DG

      And, and that are gonna get bigger.

    17. JA

      Yeah.

    18. DG

      Um, and then by the way, if you go back to those big trends that I see coming over the next 10 years, like, we don't even have the players on the board yet for those. So, um, in terms of, like, macro side, I see no limit on how big it can get. Um, in terms of our share and our ambitions as a firm, um, we probably have, like, 20% market share. Like, we measure this at least in growth land.

    19. JA

      Yeah.

    20. DG

      And, like, if you look at, like, a real frame of our market, we probably have 20% market share. Now, ideally, we would have 20%, you know, the top 20%, right? Like, we would do the 20 best of-

    21. JA

      Yeah

    22. DG

      ... the 100 things. Um, but there's kind of no limit on, like, how much more we could scale.

    23. JA

      Yeah.

    24. DG

      We just want to be doing the top 20% to 25% of things. So I think you're gonna see, like, massive macro market growth. Um, I think our share could grow a little bit, but, like, that's not a goal.

    25. JA

      Yeah.

    26. DG

      Um, I think it's more, like, we can ride the wave really well and-

    27. JA

      I mean, and even keeping up with it is gonna be hard

    28. DG

      ... keeping up with it is gonna be hard, right? Um, and so our ambitions are to be involved in the best companies across all the areas that we focus on. Um, you know, they're all technology focused, but they're pretty discreet, right? Like, there's infrastructure, there's applications, there's American dynamism, there's bio health, crypto. Like, th- those all require different things. Um, but there's gonna be new waves, and so hopefully we can, like, be part of those new waves and then ride the market growth. Um, and that's, that's kind of the sum of the goal.

    29. JA

      Yeah. Love it. David, this was great. Thanks a bunch for making the time. I really enjoyed it.

    30. DG

      Yeah, me too, man. Thanks. [upbeat music]

Episode duration: 53:11

Install uListen for AI-powered chat & search across the full episode — Get Full Transcript

Transcript of episode 4H1FbNRdtM0

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

iOSAndroidClaudeChrome