a16z Podcast - Garry Tan on Taste, Agents and Founder Ambition

Episode Date: August 12, 2026

Anish Acharya is joined by Garry Tan, President and CEO of Y Combinator, for a conversation about how AI is rewriting the startup playbook, why founders should be more ambitious than ever, and what two decades of Silicon Valley booms, busts, and missed opportunities have taught Garry about building what's next. Garry reflects on turning down an early opportunity to join Palantir, why chasing what's "hot" is often the wrong strategy, and why the best ideas tend to begin with people pursuing strange, earnest obsessions outside the mainstream. They also explore how AI changes the economics of company building, why traditional SaaS may be losing its advantage, and how tiny teams equipped with hundreds of agents can build businesses at a scale that once required entire organizations. The conversation goes deeper into agentic companies, taste and agency, why "a markdown file is an employee," and how AI could remove layers of bureaucracy that have historically limited organizations. Garry and Anish also discuss the future of consumer AI, the coming "harness wars," why AI adoption may take longer than Silicon Valley expects, and what the next generation of founders can build with intelligence that was unimaginable just a few years ago.   Resources: Follow Garry Tan on X: https://x.com/garrytan Follow Anish Acharya on X: https://x.com/illscience Follow Y Combinator on X: https://x.com/ycombinator Stay Updated:Find a16z on YouTube: YouTubeFind a16z on XFind a16z on LinkedInListen to the a16z Show on SpotifyListen to the a16z Show on Apple PodcastsFollow our host: https://twitter.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. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

We may never achieve a utopia, but it is worthy and worth it to attempt. Everything that's awesome in my life is kind of a cult. We had to go over to the Windows team. They wouldn't reply to our emails. They wouldn't fix our bugs. And we had to go over there with a baseball bat. An org like Microsoft can't. Yes.

But a startup can. Yes. And every startup must. The game has changed for founders. Gary Tan thinks their ambition should change with it. In this episode, Anisha. Chacharya hits down with why Combinator president and CEO Gary Tan to explore what it means to build a company when one person can suddenly operate like hundreds. They discuss why Gary believes founders should stop chasing what's hot and follow what they uniquely know,

how AI is changing everything from coding to company management, and why the startups of the future may look radically smaller, faster, and more ambitious than the companies that came before them. Gary also shares some of the biggest misses of his own career, including turning down an opportunity to join Palantir and why those mistakes taught him to trust direct experience over consensus. Gary, so good to have you, brother. Thank you so much for joining us today. Thanks for having me. Amazing, dude.

Well, we're going to cover a bunch of the basics, then we're going to get into the advanced topics to make sure that we kind of match the curiosity and ambition of our audience. That sounds great. Does that work? Awesome. Well, dude, maybe talk to me a bit about the beginning of your journey, journey, 2003.

It feels like what's underdiscust is founder's psychology and then sort of Silicon Valley culture and how the two are connected. What did it feel like when you first got in the game back in 2003 coming up at Stanford? I guess I wanted to work at startups, but 2003 was funny in that Web 1.0 had just ended. The NASDAQ had just crashed.

And actually, there were no jobs that I could even find in the Bay Area other than working at GAP IT. And so the two job offers I had were Microsoft for Windows Mobile, which I was pretty excited about, or Expedia. Okay. Yeah, I had student loans, and while I wanted to start a company, I just felt like I needed to get a job, and then actually there were no jobs in the base.

It was kind of hard to believe, actually, that moment was so bleak. And then right before Web 2.0. It felt like it might have been the end of tech, huh? Right, a little bit. Yeah, that was the vibe. And then at the end of the day, that's the boom, that's the bus. These things sort of happen over and over again. So at the time, did it feel high stuff?

to work in tech or on startups? Not at all. I mean, it's sort of lost to the sands of time, but I was still excited about tech, though. I had done a lot of web programming, and then one of my regrets, actually, was I was sort of chasing what was hot instead of what was clearly the thing I knew.

Mark Zuckerberg was probably doing FaceMash around then, around 2003, and then Facebook didn't even exist yet, and I was getting out of web programming at exactly the wrong time. Because it felt like that was the smart. thing to do. Yeah, it was smart. And I mean, this is true now in front of all these people at startup school. I sort of started this don't larp. Right. I saw that. But I think that's embedded in

this idea that you should just be earnest. And then you shouldn't follow what the crowd is saying, right? At that moment, people said the web was dead. But that was sort of the perfect time to work on web and social and things like that. And I didn't do it. I ran off and said, oh, well, the next hot thing will be mobile, so I wanted to work at Windows Mobile, which was true, but why I give up the thing that could have been the next thing. What I love about what you're saying, though, is often when we hear earnest, we think naive. But actually, what it sounds like you're saying is there's earnestness is a sort of result of maturity and courage. Is that? Courage is actually, I think, the right thing. Yes. If you know something, it's sort of, if someone just comes and contradicts you,

or even if like a blog post or someone on Twitter contradicts you, you're like, well, I don't agree with that because I believe my own direct experience. Yes. I know that I'm not a time traveler because if I was a time traveler, I would have sent a message back to myself and said, just stay working on web stuff. You literally have five years of experience on top of everyone else. You could do an end run around everyone.

Like you could have built any kind of social software. Yes. But instead, I was 22. So I was like being whipped around by whatever I could read in the Wall Street Journal. And the classic question I'm sure you get that all investors get all the time is what's hot and what should I work on. Right. And that's the wrong question to ask. Like the right question is what are you interested in?

What do you know uniquely? And sales, that's one of a great many mistakes I made in my career. I feel like I'm a little bit doomed to be the first principal storyteller of all the mistakes that I made. Well, I think that it's very inspiring for founders to hear your own journey and your own mistakes and then not feel so bad about the ones that we've all made along the way. Yeah. I mean, maybe talk about the culture of Silicon Valley today. On one hand, maybe it was despondent 20 years ago.

It's crazy to think that was just 23 years ago that you're talking about. Now, you think we would all be incredibly elated, yet there's this undercurrent. Maybe it's the larping thing that you mentioned. It's being outside of the light cone, the permanent underclass. Where does that come from? Do you think there's any truth in it? Talk through that. I mean, I get it.

There's just more competition than ever. And then I think maybe it's just that we're spending too much time in distribution. I think Chris Dixon talks about this all the time where you're in the IDMAs, but the cool parts of the IDMAs are like the weird fringe parts, actually. It's always what is a toy? And that's the pattern in Silicon Valley over and over again that bears repeating, right? Yes. The idea that you could have a computer on every desk in.

every home. Yeah. Today is, of course, that's what's come to pass. Yes. But the people who created that thing, they were fringe weirdo punks. Totally. Yes. They were outsiders.

I mean, Homebrew Computer Club was not where you were getting laid. Yeah. That's right. You just were like following this weird obsession that actually I want my own computer. And I think Chris has said this too. I think a good measure of how interesting the thing is can you find the edge. And for the most interesting things, you typically can't. even if you find the edge of the conversation,

you can't find the intellectual edge, which I think is a very good sign. I mean, I had this growing up in the 80s, maybe you had this too. There was sort of one way to be cool in grade school and probably in high school, and maybe when you got to college, you kind of found your people.

And now I think one of the miracles of being internet native generation is that people find their people much sooner. You find your people on Reddit, you find your people on X or Insta or whatever your network is. Oh, yeah. I mean, that's the coolest thing today.

There's probably a sub-referral Reddit, there's probably a part of X, some part of teapot that is this part of Twitter that is just your set of super weird, but really, really into that thing. And you can find your tribe more or less instantly. Yeah. Can you talk a bit about the culture of Silicon Valley? It felt like 10 years ago it was sort of, you know, it was lower status to work on something like Palantir, either because people were nervous about doing things that were too much in the national interest or because it was sort of dismissed as a services business. Now it's clearly one of the most important companies in the world

with one of the most compelling CEOs. You were early there. What did it feel like? Why did you do it? Yeah. I mean, I have a worse story than that. I have another mistake story where Joe Lonsdale and Stefan Cohen, well, my fraternity brothers at Stanford, Vassai.

Okay. And then I had a job. And they were like interns at Peter Thiel's hedge fund. Okay. And Peter has this thing where you're supposed to, you know, when you start. a thing, you write down on a piece of paper all the smartest people who you need to go and hire and get to join you. And so both of them were like, oh, let's go get Gary. He's up in

Seattle. They flew me down to have dinner with him. And Peter said, I'm so sure this is the right thing for you. Here's a check for 70 grand. That was how much I made at Microsoft at the time. Yeah. And I said, thank you very much, Mr. Thiel, but I might get promoted to level 60 this year, which I did. But I think it's like, that was like a $2 billion, $2 to $4.5.00. $4 billion mistake at this point. And I think about it all the time. Basically, I did the same thing again. It was like the same thing that got me to go work on Windows Mobile instead of just

staying on the thing that I knew and loved. It happened to get. And I was like, well, I don't know. I was working backwards from the map instead of looking down at the territory and saying, oh, well, these are the smartest people I know. Joe and Stefan were among the smartest people I'd ever met. And if not them, who was I going to be a co-founder with? The universe was saying, this is the thing you need to do.

But all I cared about was what was cool, what was hot, what would a really good investor say? And the thing is, they're like going off of a map. And you can see, you can look down, you can look around, you can have first principles knowledge of what's going on. And what they were doing is they were going to governments and three-letter agencies. And it was clear that the things that really good computer scientists in Polo Alto could do and create, D.C. did not have access to that. And it's funny how much that they had identified, like Peter and Joe and Stefan and later Alex Karp identified, that runs through almost everything that we do in Silicon Valley. The future is already here.

It's not evenly distributed. And then that's where first principles knowledge or just viewing the world as a scientist is actually very useful. Yes. And what a scientist does is, sure, there's published papers, but you go and you examine all the truths. And what you find is that once you go down into the experiment or you go down and talk to the first party people who are experiencing a thing, then you learn secret knowledge. Then you have nois. And so from Palantir, I definitely learned that everything that's awesome, in my life is kind of a cult

that starts with some sort of truth or belief that flies in the face of an orthodoxy, which is very punk, actually. Yes, yes, totally. Dude, that's very, Rick Rubin would approve. Yeah, well, actually, what I love about your saying is that there's a beautiful truth in the fact that you've talked through

some of these, maybe tragic misses, but, you know, the lessons are courage, the lessons are earnestness, right? The lessons are when your instincts are pulling you to work with a set of people. So maybe talk a little bit about, you know, we talk a lot about how Silicon Valley and the magic is it takes outsiders and turns them into insiders. And I feel like YC is the ultimate distillation of that. You know, you go from being two founders, perhaps from a foreign country to two people with YC on their LinkedIn.

And everybody now knows what that means. Totally. So talk about how that happens. Is it a process? Is it a mindset? And how has it changed over the years? I mean, I have to credit Paul Graham and Jessica Livingston and the original founders, including Trevor Blackwell. and Robert Morris.

I mean, what Paul and Jessica figured out was that there are lots and lots of really smart people who do not get access to Silicon Valley. I mean, I think in the past, you had to go to the right schools. You had to know the right people, sort of, and if you didn't, you'd come to Silicon Valley and you'd try to get invited to the right parties. There's like a social network that you had to crack. And then at YC, we have a pay. pejorative for that whole scene, which it's like we call them scenesters, actually.

Okay. Okay. Which is, I mean, it's sort of a nod to like what you used to have to do. And some people still believe that's like how you do it. All they did was say, you know what, like here's a, here's a website. Here's 12 questions. Later, we added like a one minute video. And people on the other end who are builders are going to try to give you a fair shake. And so today we have, you know, 16 equal partners, including myself. and all of us went through YC and had successful outcomes. And I think there's something valuable to that. They just welcome to Silicon Valley. We just celebrated 7,000 people coming to San Francisco. The majority of them, the very first time they came,

our startup school every single year. So incredible. We kind of call it like birthright but for tech. It looked like the Roman Coliseum, dude. It was amazing. Yeah. So, you know, I think that that's the coolest thing about Silicon Valley. Like, you sort of, it sort of doesn't matter where you're from or who you are.

It just actually matters what your ideas are and can you execute. The only other thing that I'd say, like, why see people get that a lot of people don't get is a community of people who you can be super real with. It's like, if you go to a tech crunch event, you look to the left, you look to the right. And it's like, are you really going to be able to tell them anything real? Like, you're going to hold your cards pretty close. And, you know, you're going to say how you're doing, and they're going to say killing it. And then they're going to ask back and like, oh, I'm also killing it. I mean, honestly, founders really need that.

Like, it's very solitary. It's, you know, on the day that you lose your best customer, on the day that your best engineer quits, like, who you're going to call. And if you have no one, like, I mean, the day your co-founder loses hope and you're not allowed to. Right. That's a tough day. Yeah. Talk about the historical upside sort of surprises of, like, founder types, business. types and the downside surprises.

What have you learned through YC? I mean, we're sort of in the middle of a big transformation now, I think. For the longest time, we really, really believe that you needed co-founders. And I think that everything else being equal, having co-founders is net really, really good. Totally. I think some of this is like the beginning of a cult. Yeah. Like, you know, a cult where there's only one person and that person can't convince anyone

else to join them is like, you know, worse than a cult. It's just like a crazy person actually. Right, right, right. I think this is like an old Derek Sivers TED talk, right? Where it's like, yeah, one person dancing in a field is a crazy person. Yes. But, you know, it's really the second person who like gets up. That's right.

And joins that person, like then what you find is, you know, suddenly 10 people are they're all dancing. Then it's a party. Yeah, that's literally the TED talk. Love it. And I think there's something to that. But, you know, I think. What's happening right now with vibe coding and agented coding, literally any given person could be 400 of that person.

Like from two years, like even like nine months ago, you can be a multiple of yourself. Isn't amazing? Yeah. Do you think founders are more ambitious than they've ever been? They should be. Yes. If anything, like that's the number one thing that people should do.

Yes. Is realize that, like, don't chase yourself. success or my success. Like that was, you know, when you were coming up. Yeah, that's right. The game done changed. You know, like, you actually, you can't just do a SaaS thing. Like, you know, I was just doing an analysis on like pure SaaS and I was like, oh, shoot, like, you know, a pure like per seat SaaS thing, like not totally clear it will exist in another five or ten years. Like, it's perfectly fine as a wedge. But if you're doing that now in 2026, like, you better hope that that's like a leap to some sort of moat around data or network

effects or something else, right? Yes, yes. And that's like, you know, if you, if we would time travel to like even two years ago, it's like SaaS was God. Yes, yes. Yes. The 1020x, like next 12 months revenue multiple was like, you know, iron law. Yes.

And like it's hard to believe. It was like, oh yeah, no, that's not true at all now. Well, this is, you know, a related thing that I love is that code is no longer precious. Mm-hmm. My God, it's like the things that we had to go through to get something shipped, and that's why you had to have this, like, sense that what you were making was important with a capital I. Oh, yeah. And that actually flies against the kind of beautiful truth we were just talking about,

because often earnestness isn't obviously important with a capital I, so you pursue things that seem important when most of the most important things actually start out seeming actually unimportant. Oh, yeah. I know you've been all over this with G-stack and G-Brain, and, you know, when you started out to build those things, with the intention. to like contribute to the community or just explore your interests? Oh, I was just messing around. Like, you know, what was it?

I think Boris Churny came on Light Cone podcast. Right. And I just felt like the urge to, you know, and a bunch of our mutual friends, like all of them basically just started coding again. And I was like, oh, something's happening. I better start. And then initially it was just messing around. It's like, oh, I mean, what would it be like to do and like make an office hours?

Yeah. And then the craziest thing there was like we had just started like recording all of our office hours actually. And then I could, I had like a few months of data. And there were like very clear things that the YC partners used as like rhetorical devices or ways to like draw out like a step function in your ambition. And then what's funny is I have this markdown file that is basically like the perfect script on how to advise. founders to like increase their ambition yeah and then I actually had to go into Claude and I said reduce this by 90% reduce the strength and intensity of this prompt by 90% and then I open source that because some of it's like you got to

come and that's like you guys see you know well we'll take your fable five okay thank you yeah we can't give away the crown jewels like for free this just a taste the first one's free you know well dude Kevin caught this I think it was a G-stack or G-brain, we're deep in the open source. You'd said, please apply to YC. Yeah, yeah, yeah. But only if you're a top 10% user of Cloud Code. Incredible.

Like, based on whether you're thinking about users, like thoughtful about your product and your design, like it's not everyone gets that message, actually. Yes. Well, dude, I love that you're doing it. I love that you're doing it so publicly because I think this comes back to the earnestness point, which is the whole thing about building intuition at this moment

is just using the technology. Oh, yeah. You know, and I think that there's this feeling like, oh, how can I possibly catch up, or there's so much seriousness around it? It's like, no, just use every model and build completely trivial things for no other reason to have a chassis

with which to learn the new technology. I think the most exciting thing is like, you can just have ideas. And then in the past, it's like, oh, well, I had to be the PM, let me write the spec, like, let me make sure I have, you know, one or two engineers, like, oh, how are we going to do QA? Like, there's just all this stuff that you used to have to worry about. Yeah.

And then now it's like, no, slash QA. Like, it's done. But, you know, yeah, we tested it. It's good. Like, we have unit tests, like 90% test coverage, you know? Totally. You know, it's like arranging your Zen Rock Garden. Oh, yeah.

It's just fun to do it. And the, like, the output doesn't matter that much, which is why everybody talks about coding agents, but very few people are showing their work. Oh, yeah. Because a lot of it is sort of low stakes work, and I think that's beautiful and okay. Yeah.

You know? No, I mean, and then I think it's crazy to what degree, like, suddenly agency and taste, just like off the chart's important now. Right. And then the hard part is like helping people realize like just by the act of doing it and exercising it, like that's how you get better. It's I don't actually think that like those things are innate or inborn. Like some people have more of it coming out of the womb per se.

Sure. You know, it's just like a practice. Well, let me challenge you on this because Nikita said this a couple, maybe a year ago and I thought it was interesting, which is, hey, we were told that the like highest form of art in business. was, you know, film and software. And now anyone can make films and anybody can make software, yet the same 20% who are originally doing it are still the ones who are doing it. Is there kind of this, like, is agency something

that you can, you know, embrace, or are they just people that are naturally high agency and many that are not? I mean, if I look at my own career, like, I would easily say that I'm a late bloomer. And, you know, I had to make these, like, ridiculous mistakes, like, saying no to Palantir or, like, leaving web programming to like learn. And so I feel like my life is somewhat like a testament to

no, no, like I can make mistakes and then I can introspect. I can reason about them and then make different choices. Amazing. That's the hope anyway. Yeah, that is the hope. Talk a little bit about loops. I think they're kind of coding loops, which are well discussed and now the graphs thing, which is sort of, you know, both funny and also real. But it feels like the most interesting loops are business loops. Oh, yeah. Like, how can you actually have a loop where the output is changing the way the business itself operates? I know Tom gave this great talk about how to organize your company in that way. Talk about how YC is organizing around loops and how you're advising your companies too. Yeah, absolutely. I mean, I messed around with OpenClaw. I created GBrain, which is kind of like soda

retrieval. I basically distilled earlier this year, I created gary's list.org, which is sort of like actually an agentic newsroom. And then I had to create like really, really good rag. around like, you know, personal deep research just to be able to put together some of those articles. And then all I did was like I had OpenClaw look at that repo, extract out like the agentic rag part, especially around like after the fact graph generation. And then it just ended up being soda even against like Memory Palace and a bunch of other open source. And then what I realized is that just allows the agent to know what, what million tokens to keep in its head for any given task.

Yes. And so what's interesting is like I think there are a couple things at work right now. Like the frontier model labs that are, you know, letting you use chat GPT or Claude, I think that they're still pretty constrained on cost and pretty constrained on amount of compute they want to give you. And so if you really want to token max, you actually have to use something like Hermes agent or open claw. And then you have to, like, tune it all the way up.

Like, you're just like, give me, like, let me load a million tokens or 800,000 tokens in to any given request and, like, have that be in your, like, sold on MD. But when you do that, like, I think that you basically get to live in 2028. Like, you know, it costs, I don't know, 50 or $100,000 a year to, like, use the agents at full strength, like, full 150 IQ. request for every request. But you'll feel it more or less immediately. Yeah. It just costs like a crazy amount. But for a CEO or for a founder, it actually makes a lot of sense to do that.

And you have to give yourself permission to token max in that way. But, you know, and then what you get is like I get, you get live in 2028 today. Amazing. But I think like going back to what founders can do, if you live that way, not only, do you want to do that task? You want to skillify it. You do some feet of strength, and then you turn it into a markdown file plus code plus tests

that can be reused and put into a cron job. Yes. And so I think we're just seeing across the board, a markdown file is an employee. And it's an employee that will do the job perfectly every single time. And it'll do it as many times as you want. And then at that point, all you're doing is like take any business process that you need in your company and you just do it once perfectly. And the first time you try to do it, it's going to be bad.

Expensive. Yeah. Yes, many iterations. Yeah. But then at the end, you can, as long as you can tell very, very quickly and in very simple ways, like, hey, agent, fix this, fix that. And like, this was wrong. the actual trace and the actual history out of that agent working with you, it'll actually turn that into a skill file that's perfect, actually.

And any time it screws up in a future case is just a bug fix, and then it's there for up. And this loop you would apply to sales, to marketing, to customer support, to everything. Yeah, to everything. Really interesting. I mean, GSTAC was basically that for engineering, right? Yeah. It's like there's a QA loop. I had to build something that works with brand.

browsers to go in, because that's what I found. Like early with GSTack, I found myself, you know, automating how I thought about PM. And then, you know, I had an engineering skill that was like an engine manager that made sure that there was unit test and end-to-end test coverage. But then in the end, I found myself like, I automated all the other things so much that all I was doing at the end was black box testing to verify. And I was like, oh, no, no, okay, like even that part. It's interesting, like, building these agentic systems today, I think that's all you're doing is you're seeing where the bottlenecks are. And then, you know, almost every single time,

it's like, how do I just instruct the agent to create a piece of software or a markdown file that blows away that roadblock? Right. And then when you take a step back and you've done that for the whole task, it's like, I mean, there are companies now today that, you know, we were talking about earlier. It's like literally you can go from zero to 15 mil ARR in about four months with like two or three people and like hundreds of agents, like, you know, a few hundred skill files.

It's amazing. So there's this like recursive self-improvement. There's this hill climbing that happens. Like at what point does it start to fall over? Is there like, do you need error correction? Does it air is compound? Does it get constrained by exception handling, human relationships? Because otherwise these companies should just be able to grow in.

run forever. Yeah. I mean, I guess my experience, once you get a lot of markdown files or anytime you have like some large amount of data, that's where like provenance matters. That's where conflict management sort of, you know, when you have two facts. Yes. If you have provenance, then you're like, well, this is the more recent one. Like this is the one that wins. You need a cron job that's sort of sweeping through that and making sure that things are accurate and from the right places. But these are things that I think as you use the systems, you just know that you need to build these things. Like anyone who's, I mean, this is actually one of the interesting things I think is an advantage for founders in the 30s and 40s now. Yeah. Like, I mean, I think

that there are going to be as many Patrick Collison's as ever. Yeah. But one mega trend that we're seeing is like 35, 40, 45-year-old founder who, like, been around the block. Yeah, yeah. Built a lot of engineers. I mean, I think, yeah, Peter Steinberger is, like, a perfect example of that. Whereas, like, he's been around the block. Like, he knows what to build. Yes.

And then if you take that person, like, suddenly there's 400 of those people, like, you can outperform an entire department of, like, any Mag 7. I also think that if you've had some experience in the enterprise, there are these dynamics that, you know, you can seek to kind of mirror and explore. So one of the things I'm really interested in is conflict, right? Most people at an enterprise would say conflict is a bad thing, but actually think that there's this sort of productive, constructive conflict that at its best is the search for truth. These systems allow us to do things like explore conflicting approaches and totally unbundle it from the human emotion. And I think that's actually very

important because this is why the sort of the decentralization in an organization or an economy works well because every idea gets tried. Yeah. Do you know Pedro from Brex is like the most aggressive person I've met who does this with his exact team? No. So which is cool. Like he uses like he actually was one of the people who like really got me into OpenClaw. Okay.

Because he just came in and, you know, he had lunch with us at YC. Yeah. And he just walked us through. The vibe around OpenClaught was that it was too unsafe and dangerous and like delete your whole computer and stuff like that. that what he ended up doing was like building an entire open source layer, which he actually has open source called Crabtrap. That is like an agent that watches all the network traffic and all the things that OpenClaught does.

Nice. And then that's how he squares the circle. Like Brex is one of the most regulated fintechs, of course, of course. And then the upshot of what he got is that he has his agent look at the meeting transcripts from like all of his direct reports. and he can see that like two levels down. And then the thing that really jumped out at me is like he can use meetings that he's not in to figure out like what's broken, what's in conflict, like who's in conflict with who. Yes. And then sometimes he can show up at a meeting.

He has perfect context from like the last three weeks of like that one compliance team. Yes. And then he can walk in and he's like actually you're right. We're doing it your way. And then he can leave. That's so badass. Yeah. And it's like, can you imagine like that is actually one of the defining reasons why startups or businesses fail?

Yes. It's like it's actually like a management challenge. Like the business becomes too big to fit in one person's head. And then this is like the most profound form of you have agents, you have memory, you have retrieval. You like zap the rocks more. The rocks give you like raw intelligence about like what's happening. and it can actually introspect not just on the opinion of like someone two levels down in your organization. Yes.

But it can actually look at the ground truth from the data. Yes. And it's like that, you know, that's clairvoyance, right? Like, you, I mean, if you, once you have that in your brain, like actually every business in the world is available for you to take their business. Yes. Yes. Like the world is basically built on things that are built around human beings. And human beings can only keep seven plus or minus two things in their head at any given time.

Yes. And, you know, guess what? You plus an agent can keep basically three Harry Potter books in your head. You can like rifle shot. More to come. Yeah. I mean, and more to come, right? Like this is as bad.

There are so many cryptosms that like applied like 10 times more to AI. It was like, this is the worst it will ever be. Yes, actually. Yes, yes. And so the whole world, like all business, all things that people rely on, government especially, is basically still built on a world that is predicated by human beings, limited human beings that can only keep seven plus or minus two things in their head. Yes.

And it's not ready. Like, it's not ready for the world where agents can think at the speed of light. Like, you have unlimited amounts of it, like unlimited amounts of data. And then, you know, my hope is that like basically products are going to be 10 or 100 times better. Like they should be a thousand times better. Yes. Like services should be a thousand times better. Yes.

Right? Like, you know, the invisible hand of capitalism at that point, like properly done, it's not about lowering the cost of something. Yes. It should actually be creating these like sort of magical things that I was like on Wall Street Journal podcast recently. Yeah. And the reporter, like, kept trying to, like, gotcha me. He kept saying, like, are you, you know, you guys in tech, are you trying to make a utopia? And I was like, this is a trap.

Like, I don't, I would not go so far to say, you know. But as I went through, like, a bunch of examples from, like, a bunch of the startups that you and I have, like, worked on and funded and things like that, was like, actually, like, maybe. Yeah. Right. Like, you know, we may never achieve a utopia, but it is worthy. and worth it to attempt. Right? And that, you know, I would not say

that Silicon Valley or tech will achieve a utopia. I will very earnestly say that we are trying to bring about a better world. Maybe another way to say this is just, this is the least happy you'll ever be. Yeah, you know? Because I think like one of the other implications of what you're describing is just the ergonomics

of the sort of white collar job today. You know, you think about, every, you've worked at big companies, like you think of just the kind of morass you have to get through the torture. You think of how lossy the communication is, like the entire management science sort of managerial class, like maybe that was a hundred year distraction

from what was founder mode in the Industrial Revolution, all the way through Henry Ford, all the way through some of the early Silicon Valley founders, and then everybody just kind of hit this tidal wave of management. Oh, yeah. Now we have this new model. It can lead to dramatic productivity improvements, like three days a week,

of really intentional work, maybe one day. Yeah. And the other days you're touching grass. And I think the time that you do spend is just so much less burdened by this stuff that you always felt like you had to wade through. Yeah, that's right. I mean, if you look at the typical Mag 7 day to day,

it's insane. Like I remember when I was 22 working on Windows Mobile, we're working on these crazy devices that were like years before the iPhone. Yeah. And we needed integration, for instance, from the Windows team. which was like, you know, on the other side of the highway in Redmond. Right.

And then I remember, you know, me and my sort of PM mentor, we had to go over to the Windows team. And like, they wouldn't reply to our emails. They wouldn't fix our bugs. They wouldn't even, like, mark it as won't fix. They just ignored us. Yeah. And we had to go over there with a baseball bat.

Like, I don't know why he brought one. Like, it wasn't like he was going to, like, rough anyone up. It was just like this vague threat or something. Incredible. It was actually hilarious. And I think about that all the time. It's like what kind of like bureaucratic hell is it where like two perfectly smart people who are making like good tech money. Yeah.

You know, are put into a situation where like there's a fiefdom over here. Like literally there's like seven levels between us and that person. Yes. And this person doesn't even know that we spent like four hours trying to get like a P3 bug fix that was like a blocker for. a thousand of our users or something. And how much humanity is just wasted? How many 2 a.m. sort of anxious wakeups were there?

Because somebody wanted to not have a hard conversation. Like it's, you know, it's a really, it's a waste of the way we should be spending our time and working. Yeah. And that's a profound version of like the seven plus or minus two. Like I don't think that like the, you know, product unit manager, wherever the heck their, you know, their title was. Like, I don't think that they wanted us to waste our time like that.

Yes. I think it's just that, you know, your day, like, you just don't have enough hours in the day. And we're like incredibly limited human beings. But, you know, yeah, it's funny. Like I think that, you know, I think about my time at Microsoft and it's like I remember getting an iPod and like doing competitive research for Windows mobile versus the iPod. I could imagine today at Microsoft, you could imagine a markdown file and a skill. Yeah. That is run on a Cron.

Yes. And it's like, how are all the people working on this team? Like, are they blocked? Who are they blocked by? What are the external dependencies? Like, suddenly there is a graph. Yes. There's a loop. And then suddenly, like, you know, the products can actually just move back.

Like, you could release something in a day or two that would have taken like six months. That's incredible. And if you could like multiply that out across like, the thing is like probably an org like Microsoft can't. Yes. But like a startup can. Yes. And every startup must. Yes.

be organized this way. I also love what you said. It's almost like the mid-level of bureaucracy should be agents. Yeah. Because I actually think that, you know, being in charge is somewhat overstated. Like I'd love to actually have at least the mid-level, organize the work, you know, resolve the conflicts that it can somewhat programmatically, probabilistically programmatically. Of course, the executive team maybe sets direction, the individual sort of decides on execution, but just the whole organization layer, which a lot of product managers historically did, should not be humane. I mean, the great thing. thing is, it doesn't have to be, do you know how Venkatesh Rao has this thing where it's like,

you need to be above the API line? Like if Uber came out, like if you're an Uber driver, you're below the API line, like you should be fighting to be above the API line. I actually think AI erases the API line. Yeah. Like one of the funniest things I saw recently is that there's an API company that we funded called Klaweiser and they released a bug report endpoint that is not for humans to use. It's for agents who use their software to use. And there's an LLM where it's like if there's a bug in the software

like the bug report goes, but then it replies real time. And it's like actually like that's a good bug. We filed it. We're going to fix it next week. But also like try this work around. Wow. And then like I can't get that out of my head because you could imagine

that could be how companies work. Right? Like instead of being below the API line and having to do what like, I don't know, the API tells you to do, like this can be a back and forth, actually. And then I always think about, you know, the reason why cars in the 80s got so good in Japan was it was actually a management technique. Like they realized the Toyota production system.

That's right. meant that like if you are the person doing the work on the line, you should be allowed to change the line because you are the person with the most amount of context to be a, you know, able to improve your own working situation. And that's why American cars really started sucking and why Japanese cars became absolutely incredible. There's like this very subtle, simple way that actually empowers the worker. And so there's a sort of like mix that I think about it's. Which is so interesting because what's traditionally seen as a hierarchical society came up with this very bottoms up way of working, which led you like arguably a 40-year generational run because

of process power. Yeah, which is beautiful. How many, as you think about, as you think about, kind of our evolution towards this vision, how many of these problems do you think are intelligence bound, like we need smarter models, versus there will be other attributes of models that will unlock success? I mean, honestly, it's all human. Humans are the ones who are going to, like, slow this down. And then actually, I'm realizing this might be a white pill.

Okay. So if you look at, like, the tenor of what's going on just broadly in the discourse about AI, it's all like this sort of doom and gloom, like all white collar jobs are going to go away, there's going to be a permanent underclass. Yes. And then funny enough, what I think is all of the bureaucracy

and the slowness and the seven plus or minus two of like every company, every institution, every organization, all the middle managers in the world, that is actually the white pill. 100%. So, you know, these things are going to be slower than you think. Yes. Right. I looked at, you know, we had 7,000 people. Most of them were 18 to 22 years old. Amazing. And I looked in their faces and like they're all AI native.

Like they were born more or less like, you know, they became adults in the time of chat GPT. Yes. And it's going to, you know, and for us, it's like now we, you know, run shit, which is nice. That's great. But, you know, imagine us like 20 years ago. Like, you know, we were the ones who were native to the web. You're the ones who are native to mobile. And then now we run shit. So then all of society runs on the things that we expected from when we were 18 to 22. So I think that the white pill to me is like it's going to be 20 years and that's not a bad thing. That's actually a good thing.

Yes. Like society is way slower than you think. Government is way slower than you think. Every company in the world is way slower than you think. Yeah. And there are real moats to these things. Like there are structural reasons why a Microsoft isn't going anywhere. Yes.

And then I guess this is sort of a little antithetical to what most, you know, most people in Silicon Valley would say, like, I want to start up to replace all of them. Yeah. And the reality is like, they won't, you know. They won't. I love it. Okay, I have two more questions for you. You know, one on this kind of sort of trying to envision the near future.

What do you think the next computer looks like? What does a new computer look like? Oh, man. I think in the short term, it'll probably keep looking like the way it is, but it seems inconceivable that it'll stay that form factor. Like, it's pretty clearly going to be voice. Yes. I think there's a lot around memory. It's going to take a while.

We're at this moment where everyone's sort of happy with ChatGPT or Cloud Co-Work or whatever it is. But I don't think they will be in the long term. I think you're going to want a lot more context. and you're going to want like all watched over by machines of loving grace. Yes. You're going to want this thing that is benevolent, knows about like your hopes, fears, and desires, and then it's sort of like relentlessly searching to help you in that way.

And that's actually going to be really exciting. And, you know, the path through that will probably be like computer use, memory, like, diarization, like ingestion. Yes. Yeah. I mean, I think there's going to be a lot. there and you know the cool thing is it's actually going to take longer like this isn't a like I don't think people even talking about this now my guess is like next year 2027 will be like

the harness wars yes you know and then but it's really a cost thing like it's just too expensive right now yes you know but it won't be like there will be a war for a billion consumers right when like the frontier model compute of today will be like maybe 50 or 100 bucks in like two or three years. Yeah. And that will be when like the most few, the browser wars will be back on. Yes, yes.

Well, we're already seeing it. Like the price per token of Frontier is going to infinity. Like my thesis is infinity dollars a token. You can't have it. And the price per token of the model that was Frontier a week ago is collapsing to zero. Yeah. And I actually think to me that is a huge white pill for a lot of reasons.

You know, one of the most important of which is just, we need more consumer AI software. And in a world where most consumer products are free to You just can't do that when you have really high costs of incremental distribution. The magic of software is zero cost incremental distribution. We have these magical primitives. We have computers that are emotive and have emotional texture for the first time in 40 or 50 years. What a tragedy that we haven't built more ambitious consumer products yet.

So I'm really excited for the sort of changing economics of models to unlock these new categories. I mean, and we're so early. I was convinced that becoming a convert to OpenClawn made me very, very convinced that it was going to happen right away. And then at this point, I'm like, that's not correct at all. Like, you know. Well, you should try.

The new ChatGBTGBT desktop app actually has got a voice mode. And it's sort of, it's an operator or a concierge that can see all your threads. Oh, cool. And you can access it from your phone. It's very open claw-ish. And I'm sure Peter had his hands all over it. To me, it's the closest thing we've seen yet to a true consumer assistant. Oh, I could see that.

It's awesome. Yeah, yeah. ChatGPT always had like basically the best memory. So I got to try it again. They've done a great job. Okay, the last question. You know, we talked a lot about kind of utopian visions for the world and problems that are maybe intelligence bound.

You've also put a lot of energy into sort of local politics and doing things I think are very pro-social, but it seems like more of those problems that are human coordination bound. Oh, definitely. How do you get energy to do that work? You know, is there a white pill for that aspect of our society? Maybe talk a bit about that.

I mean, the white pill is just believing that America still has rule of law and that people in government are trying to help and like somewhat earnest. Like it's not perfect. I think the white pill for me recently was even just doing jury duty in San Francisco and to see like this orderly process of really a cross-section of San Francisco. One thing I realized was like actually we are a big part of the city. Like we're 10, 20% of the people are like directly working in tech. And so to see something that was, I mean, the judge was really good.

I mean, it was a horrible case. But like, you know, government is not here to like ruin people. And so and then at that point, we can vote. Yes. Like we actually have a say. And then I think everything is downstream of media at that point. Like I think the San Francisco moment that got me very engaged was realizing, you know, Asian Americans are 25% of the voting populace,

about 30% of the population. But during COVID, like, we had a district attorney who was turning a blind eye to some of the most heinous crimes against our elders. And simultaneous to that, we had a school board that was openly hostile against Asian American kids who just wanted to learn algebra in public middle school, which hit me because I was that public school kid

in Fremont in the East Bay. Like, if I couldn't have taken algebra in middle school, middle school, I couldn't take calculus by senior year. I wouldn't have been able to study engineering at Stanford and I wouldn't be here doing any of the things I get to do today. And I'm like, oh, okay, now they've made it personal. And like, let me understand. And it turned out to be like sort of all the same people, the same people who want to destroy

all data centers, the same people who, you don't want to build any housing in anywhere in California or the world. Like it's sort of NIMBYs who want to virtue signal. who sort of have these ideologies that like don't square with mine. I mean, as a builder, you're like, well, the words have meaning, like supply and demand exists, markets exist, like incentives exist. And then, you know, I think that government is, you know, when people like us don't speak up, when we don't like pay attention

and like hold our media accountable, not just the government. Yes. Then things get very deranged. And that's what San Francisco was, I think, around COVID. And then just by sheer organization, like, you know, this is, I think that, like, the Chinese of the West Side organizing around algebra and, you know, the deaths of our elders, going to farmers markets and, like, literally physically, like, telling each other, they're killing us. They're killing us, right? But ABC News, you know, literally the mainstream media TV channel of San Francisco and the Bay Area would not cover anti-Chinese and anti-Asian crime. And, you know, it's like unconscionable, right?

Like, our institutions failed us. And the great thing is, like, when the system becomes so odious, you've got to put your body on it. I think that was Mario Savio. And it's kind of, I mean, it's ironic almost to like say that. But on the other hand, like that's what I believe. Like they can't do this to us and we're going to stop it. And even at like personal peril, even at, you know, reputational cost. Like we should be willing to take that cost because that's our covenant with our fellow man and our fellow citizen, you know.

And if we don't do it, who's going to do it? Like nobody does it. And I'm super proud of what we're doing. Francisco because we have beat back that ideological set of people who like created this mess. And, you know, it's not perfect in San Francisco, but it's getting better. It's amazing right now. I mean, I think I saw that Lurie might be the most popular mayor or politician in the entire country. Yeah.

Things have gotten so much better. Everybody sees it, feels it in the vibes. Some people say, I wonder why. But I think for those of us who know, we see that your sort of stewardship, not just of the industry of the startup community. of sort of technology, but also of the community, is one of the big catalysts. So thank you for doing that work. Thank you so much, yeah.

And I think San Francisco should be the beacon for pushing this back in every blue city in America. And that's what Gary's List.org is, basically. How do we take the lessons that we learned here? And, like, L.A. needs this. New York needs this. Minneapolis needs this. Seattle needs this.

Well, and also let's find, you know, to connect the two, the sort of, you know, culture, society, and technology, like, let's find very local ways for the ambition that this technology unlocks to show up. You know, it was in London last week, and we were talking a little about what's possible. It's like, hey, you don't have to have an American vision for how AI impacts your society. If you want the NHS to be incredible, then, like, you could do that with his technology. So the kind of the opportunity for ambition to be very localized in our communities, I think is something really interesting. Yeah. And so, yeah, act local, like, take care of the people

right next to you. Like, you know, I think it's funny. So many people get very, very obsessed with national. I mean, we want to look at national generally right now. It's like, you know what? Like, I care about San Francisco. I care about housing. I care about crime.

I care about, you know, treatment and recovery. Yeah. And if we do that, like, we can have the society we want to live in. And then, you know, if we fix local, like state and national will fix itself. Because, like, on a five or 10-year basis, like the people who win up there, like, they they will come up through a system that's functional and about saying no to things that don't make sense, about saying yes to the things that are awesome for the people.

I love it. All right, Gary. First so many. Thank you, brother. Thanks, bro. This is great. Cool. Yeah.

Thanks for listening to this episode of the A16Z podcast. If you like this episode, be sure to like, comment, subscribe, leave us a rating or review, and share it with your friends and family. For more episodes, go to YouTube, Apple Podcast, and Spotify. Follow us on X, A16Z, and subscribe to our substack at A16Z. com. Thanks again for listening, and I'll see you in the next episode. As a reminder, 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. Please note that A16Z and its affiliates may also maintain investments in the companies discussed in this podcast. For more details, including a link to our investments, please see A16Z.com forward slash disclosures.