Skip to content
a16za16z

Software Finally Eats Services - Aaron Levie

Should the US put a price on H-1B visas, or would that block the flow of new talent? Are AI coding agents actually making teams way more productive, or is it just hype? And in the AI platform shift, will the big winners be incumbents or new AI-native startups? Erik Torenberg is joined by Box co-founder and CEO Aaron Levie, a16z board partner Steven Sinofsky, and a16z general partner Martin Casado to debate the biggest questions in tech. They unpack pricing vs lottery for H-1Bs and what we’re actually optimizing for, why Box now ships a third of its code from AI, the shift from writing to reviewing code, and why bottom-up personal AI tools succeed where top-down “AI pilots” struggle. Timecodes: 0:00 Introduction 0:55 Latest immigration policy and who benefits 2:46 Salary bands as a solution for tech talent allocation 5:39 Optimizing immigration policy for wages, jobs, or merit 8:08 Market dynamics and policy changes in tech hiring 12:52 AI effects on labor productivity and developer output 19:25 Drivers of large AI productivity gains vs plateaus 24:40 Measuring AI’s impact on productivity and what’s missing 31:32 Human Taste and AI Tools 37:47 Young founders building companies differently with AI 41:34 Platform shifts: startups vs incumbents 49:01 AI opening new markets beyond software 55:54 Incumbents vs disruptors in the next decade of AI Resources: Find Aaron on X: https://x.com/levie Find Steven on X: https://x.com/stevesi Find Martin on X: https://x.com/martin_casado Find Erik on X: https://x.com/eriktorenberg Stay Updated: If you enjoyed this episode, be sure to like, subscribe, and share with your friends! Find a16z on X: https://x.com/a16z Find a16z on LinkedIn: https://www.linkedin.com/company/a16z Listen to the a16z Podcast on Spotify: https://open.spotify.com/show/5bC65RDvs3oxnLyqqvkUYX Listen to the a16z Podcast on Apple Podcasts: https://podcasts.apple.com/us/podcast/a16z-podcast/id842818711 Follow our host: https://x.com/eriktorenberg Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details, please see a16z.com/disclosures.

Aaron LevieguestSteven SinofskyguestMartin CasadoguestErik Torenberghost
Sep 24, 202559mWatch on YouTube ↗

EVERY SPOKEN WORD

  1. 0:000:55

    Introduction

    1. AL

      The universal adoption of this as a consumer technology and then bleeding into prosumer is, it exceeds anything I've ever-

    2. SS

      It's unbelievable

    3. AL

      ... I've ever experienced. And I think it is, it will just fundamentally change people's sort of daily patterns.

    4. SS

      This is all early adopters, and early adopters are very forgiving-

    5. AL

      [laughs] Yeah

    6. SS

      ... of mistakes on purpose. When something is brand new, a culture around it develops. The early internet people didn't complain that the internet was slow.

    7. AL

      Right.

    8. MC

      The more senior small teams that use AI are superhuman.

    9. AL

      Yeah, yeah.

    10. MC

      'Cause, like, they woke up-

    11. AL

      Right

    12. MC

      ... and they were all [beep] Tony Stark.

    13. AL

      Right.

    14. MC

      It's unbelievable.

    15. AL

      Right.

    16. MC

      And, like, their productivity is insane. [upbeat music]

    17. ET

      First I just wanna comment, uh, you posted in the group chat that the news around autism updates your PDoom. [laughing]

    18. AL

      [laughs] Yes. It only works if you show the image, though.

    19. ET

      Yeah. We'll, we'll, we'll tee it up on the screen.

    20. AL

      So, so, uh, you'll have to do the overlay-

    21. ET

      Yeah

    22. AL

      ... to make that make sense. But, uh, there's so many memes you can do with, uh, with that, uh, Fox News headline. So.

    23. ET

      [laughs] Exactly.

  2. 0:552:46

    Latest immigration policy and who benefits

    1. ET

      Um, first I wanna get into the immigration news. Um, re-re-

    2. AL

      Oh, you really wanna kick off just like really with the fun stuff.

    3. SS

      Let's always go in.

    4. ET

      Yeah. Like ex-

    5. AL

      Get the blood going.

    6. ET

      Exactly. Um, Martin, you had some interesting reactions.

    7. MC

      Oh, Martin. Yeah, exactly. Perfect. [laughs]

    8. ET

      Please. What were your reactions to, what you think of the policy?

    9. MC

      Well, it's interesting because it seems like any time the administration touches immigration, there's a huge outcry, knee-jerk outcry, and we saw a lot of that from VCs even. Um, but it's also very interesting that Reed Hastings, who is a classic lefty, like, and, you know, and has long been, was like, "I've been in-- doing policy for immigration for, you know, 30 years, and this is the right approach." And this is very much my thought, which is this system has been gamed for a very long time. It's very hard for startups to, to hire, uh, because of the lottery system. It's locked up by the large companies, the consultants, Amazon and Google. Um, and, like, that has to change. And I think a very reasonable way to do it is to, to set price because, you know, you've got, uh, a market, uh, [clears throat] and you've got-- you need to allocate supply. Price is a great way to do it. So I'm very, very positive on it. I comment about that and a lot of people seem to disagree, so I think it's an active discussion.

    10. AL

      Hmm. Um, well, I think-- Well, there's a couple, couple elements to this. So one is, um, first of all, Reed was ultimately responding to a thing that, that was no longer the actual policy.

    11. MC

      Yeah.

    12. AL

      So, uh, he, he said 100K a year was a great policy, and obviously the internet had moved on. Um, uh, you know, the-- it's not to me obvious... I wouldn't conclude the same outcome, uh, that, that you just concluded in that, that I, I think that you'd have a situation where the Amazons and Googles would, would probably actually capture the, the vast m- uh, portion of the talent in this situation. So it's, so it's not clear to me that like startups sort of are, come out ahead-

    13. MC

      Well, we'll-

    14. ET

      Or better off in most cases

    15. MC

      ... well, we'll step back. So, so-

    16. ET

      From, from this particular implementation

    17. MC

      ... so, so maybe, maybe,

  3. 2:465:39

    Salary bands as a solution for tech talent allocation

    1. MC

      maybe Amazon and Google who are-

    2. AL

      Yeah

    3. MC

      ... probably more easy to regulate, but there are a number of organizations that are consultancies-

    4. AL

      Yeah, 100%

    5. MC

      ... that, that actually are price sensitive-

    6. AL

      Yes

    7. MC

      ... that would be squeezed by this. And I would-

    8. AL

      Yeah

    9. MC

      ... I would think that, you know, given that they're in-

    10. AL

      Yeah

    11. MC

      ... in the top, the top 15 they make up like four or five of them, that would be a significant freeing up-

    12. AL

      Yeah

    13. MC

      ... for a higher level label.

    14. AL

      I think the-- my, my thing would be like if you could just get all the, all the people in the room that have an opinion on this topic and you, but you, you actually have sort of the, the practitioners in tech in the room as well, and the, let's say the most, um, uh, kind of, you know, you can't even say like right-wing because actually I don't think this is even classic Republican.

    15. MC

      No.

    16. AL

      Um, so, so it's just like-

    17. MC

      No, not at all

    18. AL

      ... the, the, the polls, if you got everybody in a room and you say, you sort of say, "What are we optimizing for?"

    19. MC

      Yeah.

    20. AL

      We are, are we optimizing for we don't want to have pr- like, like wages go down? Okay, that, that's a, that's an interesting thing. Are we optimizing for a particular kind of job not going to, let's say, certain populations of, of Americans? Uh, are we optimizing for just ensuring that we only have the highest merit people on the planet coming here? Like, those are all totally different kind of goals to optimize for, and I think that the, the framework you end up with and the system that you end up with should probably, you know, hopefully have like a c- like a cohesive sort of strategy behind it. My strategy would be, would be we want the absolute best in the world here.

    21. MC

      Yep.

    22. AL

      There's not exactly clear that there's a fixed number on that. Some years there might be 5,000, some years there might be 50,000.

    23. MC

      Yeah.

    24. AL

      Some years there might be 80,000.

    25. MC

      Yep.

    26. AL

      We probably want them to be net positive to wages.

    27. MC

      Yep.

    28. AL

      So let's, let's agree that like, like, you know, in any given, you know, uh, industry or locale, wages should go up with this talent pool as opposed to down. So I think that's actually totally reasonable. And you, so you should add the market kind of, you know, s- you know, sort of some market dynamic to that. And, and you shouldn't be able to kinda game and exploit the talent pools for saying, you know, "Now in Detroit we can go wipe out IT jobs because we can go and offshore those." Like, like, I think you could build a system that, that basically meets all of those goals while still ensuring that you can get somebody that goes to, you know, their master's program and, and name your, name your, uh, your state school. They come out of it, they're an AI engineer. They're not yet at the sort of Meta is gonna pay them $100 million, um, uh, but they are gonna be totally valuable contributors to our economy. It's all sort of positive sum. It's not taking a job from anybody else. It makes us more competitive. And I think there's a way to do that without sort of overly, uh, let's say putting constraints in the system that make it maybe so a startup wouldn't be able to, you know, kind of economically viably participate in this. And I think 100K per year would, would be at a point where the, you know, startups would be directly impacted.

    29. MC

      Uh, working with a lot of startups, I'm not sure that's the-

    30. AL

      The, re-respectfully, respectfully, the kind of startups that Andreessen Horowitz sees aren't, are not all of the base of startups in the world.

  4. 5:398:08

    Optimizing immigration policy for wages, jobs, or merit

    1. MC

      you could quibble about the number.

    2. AL

      Yeah.

    3. MC

      Is it 20K, which Keith Rabois said when I talked to him was very sensible, or is it 100K? I don't know. But like the idea that you get-

    4. AL

      Wait, Keith threw out 20?

    5. MC

      Keith threw out 20.

    6. AL

      Well, let's just go with Keith's number. I think that-

    7. MC

      [laughs]

    8. AL

      Like if Keith, if Keith threw out 20, I think we can be good with the Keith number on this one.

    9. SS

      But I, I think, I, I think the, the, the number, it's easy to fixate on the number.

    10. AL

      Yeah.

    11. SS

      But, but the, the real, the real... You have to always look at what is the number replacing, and I, I don't think the average person having this debate, other than the people-... that really work at this have any idea the amount of productivity that is lost working this system.

    12. MC

      Yes.

    13. SS

      I mean, the, the-

    14. MC

      100%

    15. SS

      ... the incredible amount of resources. And, and of course, the bigger companies, the, the ones that you mentioned-

    16. MC

      Yeah

    17. SS

      ... have enormous teams that spend all of their energy-

    18. MC

      Yeah

    19. SS

      ... like, literally, essentially as lobbyists-

    20. MC

      Yeah

    21. SS

      ... working the system, and then the back end of that is, are all the justifications and all of the management-

    22. MC

      Oh, yeah

    23. SS

      ... all of the hamlet.

    24. MC

      This is, this is broken.

    25. SS

      And now they just deployed-

    26. MC

      This is broken

    27. SS

      ... all their resources to manage, like, in-house call centers to deal with-

    28. MC

      Yeah

    29. SS

      ... getting their employees back to the United States to just deal with it. I do-- You, you hit on one point that I think is really important to this debate that is sort of getting lost, which is the, there's no doubt that within the, the big tech world that they, for the past 25 years or so, they really went on this sort of bifurcated curve, which is hiring o- you know, for the people in the office focusing on, say, 25 or 30 university departments. And then basically, everybody else was like, "Well, it's so much easier if we just hire huge numbers of people from these eight international locations and schools."

    30. MC

      Yeah.

  5. 8:0812:52

    Market dynamics and policy changes in tech hiring

    1. MC

      to, like, Florida today and you try and get, like, an IT job for 100K, you just can't, right? And so that, I think, is actually the area that's the most directly impacted by the large consultants.

    2. SS

      Meaning there aren't jobs that pay 100K or that you just can't find a job?

    3. MC

      They, they're taken. They're all, they're all taken.

    4. SS

      They're all taken.

    5. MC

      They just don't exist.

    6. SS

      There's no-

    7. MC

      Like, like any sort of IT administrator-

    8. SS

      Right

    9. MC

      ... like the, like the, you know, services, like the basic consulting gigs, like all of that has been saturated. It's very, very tough to get a job between, like, 80 and 120K in, in much of the United States because of this, right? And so this isn't about, like, a new grad being a software engineer, 'cause the reality is, is the expected value of a software engineer over their lifetime is, is high enough that I think that, like, the market kind of navigates that. But it's almost these kind of, you know, lower level, more, like, IT admin jobs that have been squeezed out. And listen, if we want to bring them back, then I do think, and we don't wanna do this kind of arbitrage that a lot of these companies are doing, that we're gonna have to-

    10. SS

      So-

    11. MC

      ... change the pricing.

    12. SS

      So but, but don't, but, uh, the w- wouldn't a minimum salary band effectively solve that problem for you?

    13. MC

      Sure. Yeah, yeah.

    14. SS

      Okay.

    15. MC

      For sure. Yeah, I think there's a lot of mechanisms to do it.

    16. SS

      Oh, okay. Okay.

    17. MC

      I, I actually agree with you.

    18. SS

      Yeah.

    19. MC

      Like, we should actually talk about the problem that we're trying to solve.

    20. SS

      Yeah.

    21. MC

      So I think we would all agree if you come to the United States, you get a college degree, you should have a visa, right?

    22. SS

      Yeah. Yeah, yeah.

    23. MC

      Kind of, you know, okay. So, like, that's for sure.

    24. SS

      Well, do we, do we all agree on that?

    25. MC

      Okay, yes, I agree. [laughs]

    26. SS

      Okay. So the, but like, as in, like, I don't think that the people proposing this strategy actually agree on that.

    27. MC

      Uh, well, so Trump famously said this.

    28. SS

      Oh, he famously said a lot of things. He, for sure. [laughs]

    29. MC

      [laughs]

    30. SS

      And then he unsaid it.

  6. 12:5219:25

    AI effects on labor productivity and developer output

    1. SS

      the tricky part.

    2. MC

      Yep, yep.

    3. ET

      I wanna segue from, uh, labor markets to-

    4. SS

      [laughing]

    5. MC

      [laughing]

    6. AL

      What's the next, uh, what's the next really interesting political topic that we wanna engage in?

    7. ET

      Yeah, exactly. Um, from labor markets to labor productivity with AI. O- offline, Aaron, we were talking about the Meter paper, and it was, the paper s- suggested that their developers were actually less productive with AI, but that doesn't square with your experience talking to a lot of different startups, seeing a lot of different startups and how they're so much more-

    8. AL

      Yeah

    9. ET

      ... more productive. So why don't you talk about where, where you're seeing startups say they're more productive and wh- why is it happening?

    10. AL

      Yeah. I'll, so I'll, I'll first just represent our own case study-

    11. ET

      Sure

    12. AL

      ... and then, and then there's the really extreme version. So, uh, our own case study is we've adopted a, a, a few different kind of, um, AI coding tools, um, uh, you know, Cursor being a, a, a super, super popular one internally. And I, you know, as I talk to people, let's say in the hallway, who have, you know, yeah, maybe they're trying to get me excited by AI-

    13. ET

      [chuckles]

    14. AL

      ... but like I think they know I'm, I'm bought in. So, so the, the kind of qualitative answers I get from, from people, and then I'll give you our, our internal metric, you know, some, some people say, "I'm getting, you know, a 20 to 30% productivity gain." Other people will say 75%. Um, interestingly, I have not been able to pinpoint the demographic difference, uh, on the answers.

    15. MC

      Oh, but this is self-reporting.

    16. AL

      This is self-reporting. Um, uh-

    17. MC

      How, how happy are you? [chuckles]

    18. AL

      Uh, yeah. [laughing] So, so, so... No, no, but we have internal metrics as well. So about 30% of our code right now is, is coming from AI.

    19. MC

      Oh, I see.

    20. AL

      So, so, so we've got the-

    21. MC

      70%?

    22. AL

      30%.

    23. MC

      Oh, 30%.

    24. AL

      30%. So we have some of the, uh, kind of pure internal, uh, metrics that, that show this. Um, but what's interesting is, is that, like I have a two by two of I have senior people that are saying that they're, you know, getting 75% productivity. I have junior people that are saying they're getting 75%, and then vice versa, um, on the 25%, let's say. And it's... I, I, I haven't been able to quite figure out a pattern, maybe except for, and we've talked about this a little bit, but you kinda see this online, except for maybe the, the biggest criteria is just the people that actually push the AI to do more, which is sort of this other new kinda psychographic, which is just like, who is willing to just be like, "You know what? I'm gonna YOLO this, this task-

    25. MC

      Yeah

    26. AL

      ... and just see what the AI comes up with." And your, your sort of willingness to just do that, I think probably somewhat, uh, then, then shows up in the, uh, in, in the, the ultimate productivity gain. So that's, that's us as a lar- relatively larger company, uh, on the startup side. But what's crazy, and this is the thing that just blows my mind, I will regularly talk to three, five, 10-person startup founders that, that self-report they might be getting somewhere on the order of like three to five to 10X, uh, productivity improvements. And, and the, the, the big difference is, is that, you know, a year ago if we were to have this conversation, the co- the conversation would be about, you know, AI sort of doing type ahead and, um, and it, you know, it can add maybe like a few lines of code to your productivity per, you know, incremental, you know, sort of unit of work that you give it. And then now obviously the big phenomenon is background agents where I give it a very, you know, detailed prompt. I send it off. It comes back. You know, people talk about it as like a slot machine of like some percent of the time it's not gonna come back with the right thing. You have to decide which, what you actually, you know, kinda pull in from it. But the kinda startups that are getting like real multiples of productivity gain are, are just, they're fundamentally engineering in a different way. They're sending off a task. The task goes off, comes back in 20 minutes, and then they're really in the, in the business of doing code review, not code writing, and it's gonna obviously change, you know, uh, quite a bit of what computer science looks like in the future. And then the only question is like, you know, what are all the things that that's good for? Where does that break down? What, what kinda teams can actually evolve to that state? But that one has been blowing my mind the most recently, and I think that kinda fundamentally changes what, what the future of s- you know, kinda engineering looks like.

    27. SS

      I think what you said is super interesting, but let me ask you, I, I think that there's an overlay that goes beyond junior, senior and, and there, and which is, is we're all talking about, uh, characteristics that have two, a solution that has two really important characteristics right now. One is that it's engineers doing stuff for engineers and, and they understand the domain super, super well.

    28. AL

      Yeah.

    29. SS

      And I think that that's an really, really important part and a really big thing that people aren't talking enough about, which is maybe what's going on is that you have AI accelerating for people that work in the domain-

    30. AL

      Yes

  7. 19:2524:40

    Drivers of large AI productivity gains vs plateaus

    1. MC

      of this that makes it very difficult to measure. One, one of them is, and I don't think it's just an early adopter thing, like, these, these models are so magic that you get dazzled.

    2. AL

      Oh. [laughs] Yeah.

    3. MC

      So even, even if it's not what you want, you're like, "It was great," you know? And I, I think it's very easy to conflate that with being productive. Like, "It's not what I wanted, but it was amazing-

    4. AL

      [laughs]

    5. MC

      ... so therefore, therefore it must be-"

    6. AL

      It's still doing great.

    7. MC

      But no, [laughs] seriously, therefore it must be great. So, so, like, maybe over time we just abdicate having an opinion, and, like, the model does everything.

    8. AL

      Mm.

    9. MC

      But right now, and I see this a lot, people are s- like, they're so enthusiastic about using AI, but it really hasn't impacted, you know, their output. They're just enthusiastic. The second one is I, I feel like there's almost shadow productivity-

    10. AL

      Whoa, sorry. How, how would you, uh, how would you, how would you verify that, uh, with the, a five or 10-person company who, who kind of empirically is operating at, like, a 50 to 100-person company? Like, like, you just-

    11. MC

      Oh.

    12. AL

      You can see the, the sheer s-

    13. MC

      Oh

    14. AL

      ... you know, scale of their code, and you're like, "Okay, you could not have done that 10 years ago."

    15. MC

      Oh. I actually agree with everything-

    16. AL

      Yeah, yeah, yeah

    17. MC

      ... that Steve is saying, which is, so listen, anec- anec- this is anecdotally. Anecdotally, I've worked with a lot of companies. Anecdotally, the more senior small teams that use AI are superhuman.

    18. AL

      Yeah, yeah.

    19. MC

      It's like they woke up-

    20. AL

      Right

    21. MC

      ... and they were all fucking Tony Stark.

    22. AL

      Right.

    23. MC

      It is unbelievable.

    24. AL

      Right.

    25. MC

      And, like, their productivity is insane.

    26. AL

      Yeah.

    27. MC

      But they're all, they're all super senior-

    28. SS

      And, you know, and look, they were s- they don't... I, I don't wanna take anything at all away, but those companies were also incredibly productive relative to a 10-person team-

    29. AL

      Yeah, 100%. Yeah

    30. SS

      ... at a big company.

  8. 24:4031:32

    Measuring AI’s impact on productivity and what’s missing

    1. MC

      AI productivity is hard to measure for two reasons. The first one I just mentioned, it's just really dazzling, so I think, like, people kind of like-

    2. SS

      Yeah, yeah, yeah

    3. MC

      ... they're like, "Oh, it's amazing." The second one is I think a lot of the productivity's actually hidden, and people measure the wrong thing, right?

    4. SS

      Mm.

    5. MC

      So get-

    6. SS

      Shocking that people measure the wrong thing in productivity.

    7. MC

      That's right. That's right. [laughs]

    8. SS

      Literally the history of productivity measurement.

    9. MC

      Well, but, but also what happens here is, like, you have the board, and the board is like, "We need more AI." And then, so what happens? Then they go to, like, some CT or some innovations lab, and then the innovations lab do AI, and so, like, whatever. They, like, bring in, like, build some internal tool, and it'll fail. Of course that'll fail, right? But the reality is, is, like, this AI wave is so personal. Like, probably most people in the company are using ChatGPT. Probably there is, you know, some personal assistant. Probably, you know, they're using Cursor or some coding thing, and that's much, much harder to measure-

    10. SS

      Yeah, right

    11. MC

      ... just because-

    12. SS

      Yeah

    13. MC

      ... you know, it's not advertised. And so if you actually look at the reports on, like, enterprise things fail, you guys look at, like, what they were-

    14. SS

      Right

    15. MC

      ... they're measuring.

    16. SS

      Well-

    17. MC

      It's like, yeah, clearly some internal project pushed down-

    18. SS

      Yes

    19. MC

      ... by the board where they hired, like, you know, some consultant to do it is gonna fail. Th-those always fail.

    20. SS

      Yeah.

    21. MC

      But that's actually not what's going on.

    22. SS

      Right.

    23. MC

      Like, the movement that's happening is a very secular thing.

    24. SS

      Well, so it's, it, this is, uh, this is the, the, the next time that bottom-up a-adoption is really changing-

    25. MC

      Yeah

    26. SS

      ... the productivity equation.

    27. MC

      Yeah.

    28. SS

      And that's a thing that, that it defies. Big companies do not know how to deal with that.

    29. MC

      Yeah.

    30. SS

      Because they, they want, they need to control it. They worry about safety and security and privacy and all of their corporate rules. And, and then also, the other thing I think to overlay on that is AI is, is a very unique, if that's not a bad way to say things, i-innovation in that it's, it's, like, non-deterministic.

  9. 31:3237:47

    Human Taste and AI Tools

    1. AL

      into a single action, and so we're-- it's just like a completely different way of, of thinking about work.

    2. SS

      Where, where does that fit in for you in what we were talking about, what I was asking about earlier, which is w- how does the expertise, your expertise really contribute to that?

    3. AL

      Yeah.

    4. SS

      And in particular, I, I think it'd be interesting for people to understand, like when you talk to customers, how do you help them to avoid trying to get people to make AI make them do jobs they couldn't do in the first place?

    5. AL

      Yeah.

    6. SS

      And like how does... 'Cause that's a easy point of failure.

    7. AL

      Um, yeah, I mean, it actually is... Th-this is, this is this really, um, counterintuitive thing where, and you've talked about specialization on the last one, is like, it-- the, the, the, uh, biggest gains of AI go to people who have some degree of expertise in an area to know what is actually true, what is not gonna work, what should I, what should I integrate from the output of this AI.

    8. SS

      Yeah.

    9. AL

      You know, um, like what are the 2% of things that maybe are hallucinations or, or, you know, took the data in the wrong direction. If you don't have a deep understanding of your particular space or field or domain, you, you aren't able to then have the right judgment to make all of those decisions. So I think the experts just get more powerful in this world. And so I would... That-that's why I'm not even like convinced that you can tell a college student to learn anything different than ever any other, you know, period in history. Like, like be really good at a particular field-

    10. SS

      Yeah

    11. AL

      ... and then AI is merely a turbocharger of your capability in that particular field. But like if I didn't know, if I didn't just like generally know the things I know about, you know, SaaS, which is like obviously like a really like, like weird expertise, um, but like I'm like okay at understanding SaaS, then I, then the things I give to, you know, a deep research agent that I then go and incorporate back into work wouldn't make sense like to me. I wouldn't have the, I wouldn't have all the context for like that one thing that it mentioned, how do I like form that into the overall strategy? But because I have some understanding of, of this particular industry, that just makes me way more productive. So, so I don't think a-expertise goes away at all, and I think any, any of the experts in their particular area just become more powerful.

    12. MC

      So we, we actually have a fair bit of anecdotal market data on this, so it's very, very interesting. So if you take like a lot of these, let's just take kind of a non-text example like image or video, and if you look at the customer base for any of the popular platforms-

    13. AL

      Yeah, yeah

    14. MC

      ... very interesting. So if you, if you draw a dollar at random that's monetized, it's from a professional.

    15. AL

      Mm.

    16. MC

      And for obvious reasons because, you know, like y- you know, they can produce... If you draw a user at random, it's, it's casual-

    17. AL

      Uh-huh

    18. MC

      ... and it's in the tail.

    19. AL

      Right.

    20. MC

      And so it's fairly clear that, you know, this is a prosumer movement from monetization. And, you know, I'm in, I'm associated with a number of companies that work with, like say, professional designers or professional creatives. They spend just as much time on the AI tools as they would on traditional tools.

    21. AL

      Yeah.

    22. MC

      It just turns out the out-output is far more rich and like, you know, it, you know, it, it's, it, it tends to be-

    23. AL

      They have a, they have a taste that is, that is-

    24. MC

      Yeah. But I mean, it's still, still human taste.

    25. AL

      Yeah.

    26. MC

      There's still like very specific requirements. And so I think, I think, you know, if there ever is an ads model that ever shows up for AI, I think that there's gonna be a long tail of people that wanna use AI, but they actually don't have the financial incentive to do it-

    27. AL

      Mm-hmm

    28. MC

      ... or it isn't tied to like, you know, their actual job. And we're already starting to see that bifurcate out, and there's gonna be another subsection that do. And, and my sense is, let's say if I'm, let's say, say I'm writing like, whatever. I'm a, I'm a casual developer and I'm writing a 3D game, and I wanna have like a 3D asset. I've got one of two choices. I can have AI create it for me-

    29. AL

      Mm-hmm

    30. MC

      ... or I can, you know, contract a professional to do it. Like me as a, a developer, I'm not gonna g- create... Even if I use AI, it's not gonna be a great 3D asset, [chuckles] right?

  10. 37:4741:34

    Young founders building companies differently with AI

    1. ET

      back to a, a point you made earlier about that there are 20-year-olds who are building companies in new ways.

    2. AL

      Yeah.

    3. ET

      Because remember a few years ago, I think Patrick Collison and a few others were asking, "Hey, where are all the Gen Z super successful founders?"

    4. AL

      [laughs]

    5. ET

      Remember that?

    6. AL

      Yeah.

    7. ET

      And, and of course there was Dylan Field and Alexander Wang.

    8. AL

      Yeah.

    9. ET

      But they, their companies took a few years-

    10. AL

      Yeah

    11. ET

      ... to really... But, but now, you know, we're seeing, uh, the Cursor founders, the Mercor founders sort of, you know, get to massive scale in a very short period of time. And maybe it was the, the foundation model companies required, you know, a certain level of, uh, you know, experienced founder because of the fundraising amounts and maybe the applications are, you know, m- more conducive to younger founders. But what, what's your s- re- reflection on this?

    12. AL

      Well, um, the, the, I don't remember exactly the date at which he, he mentioned that, but the, but I do think there was a period, um, between in, in the sort of, you know, mid 2010s to late twen- to early 2020s where, where we were actually in kind of a bit of a lull as an industry.

    13. ET

      Yeah.

    14. AL

      And-

    15. SS

      Yeah, yeah

    16. AL

      ... the, the reason for that was, like, it, like we kind of did, like, check off a lot of boxes of, like, the core things that people needed in the world. And so we, like, checked off, like, a lot of the... Like, once you have Slack, you don't need five other chat tools. Uh, once you have Zoom, you don't need five other video conferencing tools. And so it gets kinda derivative, you know, past these kinda core platforms. And so once you had, like, SaaS, you know, kind of check off all the major, like, things you do at work, and then in the consumer world, we, like, we had ways of delivering food and listening to music and watching videos. So, like, there's, like, not an infinite set of things that we, we do as consumers. Then, then what, what is the 20-year-old founder supposed to, to work on? Like, they're gonna... The, it's like you have, you know, pretty finite opportunities as compared to in the mid 2000s, let's say, the whole world was open. You could start anything. And because every single category had to be reinvented-

    17. ET

      Yeah

    18. AL

      ... kinda post mobile po- post, you know, kind of cloud maturity. So we now have that era in AI, and that is why I'm, like, so unbelievably pumped up. And, um, and it's because you have a complete reset of the landscape where there, there's, like, on- There, there's incumbent advantage in distribution, but that is it. There's no o- other, other real advantage that an incumbent has.

    19. SS

      Well, there's a bunch of disadvantages.

    20. AL

      Yes. And then there's a bunch of disadvantages.

    21. SS

      But I think-

    22. AL

      So-

    23. SS

      Yeah, go ahead. Sorry.

    24. AL

      No, no, no. But, like, I mean, you know where I'm going. So, like, but, like, so you have this, you have the exact makings of a landscape where, where new startups can come in and do things that incumbents either can't or there's no obvious incumbent to even do that thing. Because again, you're taking maybe, like, services and turning them into AI labor, and there was no software incumbent previously to even attempt to do that. And then you have incumbents that have a whole lot of complexity in terms of their ability to go and execute in some of these spaces, and they're not gonna retool their entire internal engineering workflows to move at 10X the pace. And so a brand new startup can go and do that and then instantly get the scale of a larger company. So it's the first time in history where you have none of the disadvantages of a big company, and the, the traditional advantage you have as a big company is you have scale and you have distribution. Scale because you can look at a feature and you say, "We're gonna go build that next month." And obviously it's harder because there's, like, you know, um, you know, the, there's, there's just lots of complexity to that. But at least you have the, the human, you know, power to go do that. Now, as a startup, you instantly have scale because, you know, background agents, et cetera. And so then it's a distribution game, and a lot of these, you know, pieces of software can go viral now in a way that wasn't possible 10 or 15 years ago. So we've kinda neutralized a lot of the incumbent advantages, and so thus it's a ripe opportunity for brand new startups. Often will be people just, like, coming right out of college saying, "Hey, it's my first time building a company." Like, w- like, they're crazy enough to not know how hard it is, so they'll jump right into markets that otherwise we would assume are like, it's al- like, the, the market's already solved for. There's no way that you're gonna build a company, and you'll just have new startups that actually go and do it and, and actually produce real, you know, real companies in these spaces.

    25. SS

      Yeah. 'Cause I, I mean,

  11. 41:3449:01

    Platform shifts: startups vs incumbents

    1. SS

      this is just so critical because it, what, what's really happening is this is why you know it's an actual platform shift. So Silicon Valley's seen this movie many times before, and it, that's why often there's a lot of this, you know, is this crying wolf or not? Because everybody knows-

    2. AL

      [laughs]

    3. SS

      ... that when there's a platform shift, that's the moment in time that, that startups are at-

    4. AL

      Right

    5. SS

      ... at an advantage.

    6. AL

      Yes.

    7. SS

      And, and so each time there's a platform shift, like, everybody's like, "Oh, this is it. This is gonna reinvent everything." And then it doesn't, and people get really like, "Oh, it's always incumbents."

    8. AL

      [laughs]

    9. SS

      But historically, like, the advantages to incumbents are wildly overestimated.

    10. AL

      Mm-hmm.

    11. SS

      And really, I, I mean, this is this one where, you know, like, you know, was, did the internet undo Microsoft or not undo Microsoft is a super interesting thing.

    12. AL

      Yeah.

    13. SS

      'Cause of course there's a $3 trillion company now, but not on the internet in a way that you think about the internet. Like, none of the consumers, none of the platforms-

    14. AL

      Yep

    15. SS

      ... none of the assets that we had in the '90s-

    16. AL

      Yeah

    17. SS

      ... became internet assets.

    18. AL

      Yeah.

    19. SS

      I mean, even if you look at Azure today, it's an amazing accomplishment. It's not running Windows anywhere.

    20. AL

      Right.

    21. ET

      Yeah.

    22. SS

      And, and I think that, that that's why, you know, it's not crazy to go, "Wow, is this gonna be good or bad for Google?"

    23. AL

      Right.

    24. SS

      Because there's a bunch of stuff-

    25. ET

      Yeah

    26. SS

      ... that becomes really, really difficult if you, if you don't make transition. And then it turns out historically, even if you do make the transition, you really didn't, and you just have to wait for time to pass.

    27. AL

      [laughs] Yeah.

    28. SS

      Well, and, and this-

    29. AL

      Yeah.

    30. SS

      No, this is like Intel-

  12. 49:0155:54

    AI opening new markets beyond software

    1. AL

      that, the, but the other thing that I just don't think we've had, at least I don't know of a modern kind of case study for, is, is again this, uh, this opening up of non-software TAM for software.

    2. SS

      Yeah, yeah.

    3. AL

      So it's not even incumbents in the classic sense. The incumbents are, are, are really just professional services categories of work. And so, a- and so it's really, for the first time ever, you're packaging up intelligence for a particular domain and workflow.

    4. MC

      Yeah.

    5. AL

      And so there's no software company you're competing against for those dollars.

    6. MC

      Yeah, but, but it, but it could be the vertical company, right? Like-

    7. AL

      Yes.

    8. SS

      Yeah.

    9. MC

      You know, like you have to become an ag company.

    10. AL

      But, but then, but the vertical company will probably also be your customer though.

    11. MC

      You have to become a construction company.

    12. SS

      Yeah, yeah.

    13. AL

      But they're also your customer, so it's actually this amazing thing where, where the people you're probably disrupting on paper are actually the primary users of your technology.

    14. MC

      That could actually take advantage.

    15. AL

      Yes.

    16. MC

      Yeah.

    17. AL

      And so then it's like, it's like there's really no inherent competition until, until, you know, eventually like more companies flood that space to do that idea.

    18. MC

      So this plays out in practice. If you have a company, an AI company that goes after, say like agriculture or construction, they end up like realizing the competitive set are agriculture and construction. The buyer knows how to price things like agriculture and construction. They end up becoming basically agriculture and construction companies, and then they end up doing exactly what you're saying, is selling to the agriculture. [laughs]

    19. AL

      Yeah, yeah. Yeah.

    20. MC

      'Cause it's like, you know, 'cause they're not good at that, right? That's not what they do.

    21. SS

      There, there's a whole world.

    22. MC

      That's exactly right.

    23. SS

      In fact, the, the earliest PC software-

    24. MC

      That's exactly right

    25. SS

      ... was extremely vertical. Like the, if you actually look at the TRS-80 catalog from the early 1980s or nine- late 1970s, it would be like, "This is crop rotation software."

    26. AL

      [laughs]

    27. MC

      This is like-

    28. SS

      Like literally like, okay, this is what you should do. Like, and you, and the, the salesperson for Tandy-

    29. AL

      That's amazing

    30. SS

      ... would show up in Nebraska and sell crop rotation software.

  13. 55:5459:33

    Incumbents vs disruptors in the next decade of AI

    1. ET

      When, when we look at mobile, there were big companies built, you know, uh, like Uber and, and WhatsApp and Instagram and TikTok, but the biggest beneficiaries were Facebook and, and, and, and Google. Um, in AI, do we think it will be different, that sort of the biggest companies in, in the world in 10 to 20 years from now will be, will have created-

    2. SS

      Mm-hmm

    3. ET

      ... you know, after ChatGPT, or, um, or will it be similar that-

    4. AL

      What's your timeframe?

    5. ET

      Uh, you know, post-2019. I don't know. Um-

    6. AL

      No, no, no. How many, 10 to 20 years you said?

    7. ET

      Oh, yeah. Sure.

    8. AL

      Okay. So we can't know if we're wrong for until we do this podcast in 10 years.

    9. ET

      [laughs] Yes.

    10. SS

      Okay.

    11. AL

      Um, I think, I... This is so boring, but I think, I think it's gonna look like what we saw in something like SaaS or cloud, which is the incumbents get bigger, but then there's all of these new categories that we would not have been able to predict, and then there's lots of 10 and 20 and 50 and $100 billion companies that also emerge, and then over time those will just continue to scale.

    12. ET

      So similar to mobile?

    13. MC

      And, and some-

    14. AL

      Yeah

    15. MC

      ... and some don't make the transition.

    16. AL

      Yeah. Uh, yeah.

    17. MC

      Like, you know, some don't make the transition.

    18. AL

      Some will go down on a relative basis 'cause their mar- or, like, their market wasn't as ripe for agentic kinda workflows. But I think that you can kinda say, um, you know, maybe this is then for another conversation, is just, like, if you have a current system of record that, that has a set of workflows on it where agents make sense to make that workflow much more powerful, that's a good position to be in. But it t- I bet you that if we look back in 10 or 20 years from now, the vast majority of things agents do don't relate to just those things that we are currently looking at because there's just so many-

    19. SS

      Yeah

    20. AL

      ... more fields that are now open. And so all of those use cases I would favor the disruptor and, or the insurgent. And then in the, in the today's spaces I would kinda favor the incumbent on the margin, but the markets are so large that all... you're gonna see kinda growth in all of them.

    21. SS

      The, th-there's a key, um, attribute across all of those which, you know, y- is, is sort of like thought leadership or, like, who, who is really setting the agenda for what people are talking about, and I think that's the thing that really changes. The incumbents become bigger, but nobody's wakes up in the morning wondering what they're up to.

    22. AL

      Right.

    23. SS

      Nobody starts to wonder, "Well, if they're gonna do it, we need to understand it." And that's the shift, and you can think of it in the enterprise space or the business space, like what do the CIOs, who do they wake up thinking about?

    24. AL

      Yeah.

    25. SS

      And that, that was a huge shift that sorta goes under the radar. And in the consumer space, it just, like, it becomes the, "I understand. I use ChatGPT at school. I, I need ChatGPT."

    26. AL

      Yeah.

    27. SS

      And y- there's nothing you can do about it as a company.

    28. MC

      Yeah. I think the more provocative question is, is are there any laggards that will use this to get ahead? And we've seen this in the past, right? Like, will Cisco do something interesting?

    29. SS

      Yeah, yeah.

    30. MC

      Oracle is making some kinda crazy moves. Like, are we gonna see those that, like, missed, like, social-

Episode duration: 59:34

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

Transcript of episode JI_lyrkh4Gs

Get more out of YouTube videos.

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

Add to Chrome