Sam Altman in conversation with Patrick Collison

Transcript

Please welcome back Stripe co-founder and CEO Patrick Collison. Good evening, folks. I really hope you've enjoyed the announcements and the sessions over the course of the day thus far. So I'm extremely excited about this interview. As you know, I was due to be interviewing Greg Brockman, who was a very early Stripe employee and then a co-founder of OpenAI. But AI is a dynamic space, and there's a lot happening. And so we've made a slight substitution. And so instead, I'm going to be interviewing somebody that I've known for a long time. I was just counting backstage. I think I've known him for 18, maybe 19 years. He was actually Stripe's one of the very earliest investors in Stripe. I think maybe Stripe's second or third investor. And he also co-founded OpenAI in 2015. So please welcome to the stage, Sam Altman. Thank you. Thank you. [SPEAKER_01] All right. I think we have some Codex fans in the audience. Love to hear that. How's the week going? [SPEAKER_02] So fun. [SPEAKER_02] It's a busy week. But I'm happy to be here. This is an unexpected surprise. [SPEAKER_01] Well, thank you for joining us. [SPEAKER_01] We appreciate it. So we opened this morning by saying that it is, we've arbitrarily decided that the singularity started on January 1st, and thus today is day 119. What do you think of that? [SPEAKER_02] It does feel like we are somehow in the takeoff. [SPEAKER_02] Day 119 feels like a reasonable enough guess. [SPEAKER_02] Yeah, I won't fight it. [SPEAKER_01] Did you feel like this? [SPEAKER_01] So we've started to see a bunch of our metrics inflect as of late last year, beginning of this year. [SPEAKER_01] I mean, things were doing well, but somehow they really just, the shapes of the curves changed. They really went parabolic. Is that matched in what you guys see? Was there some trajectory change? Why are we seeing this? [SPEAKER_02] I do think the models got really good, especially for coding, but really good in general, starting late last year, very early this year. [SPEAKER_02] And at least in my own experience of using this technology and seeing what other people are doing with it, and also this sense that every week is now a little bit different than the week before. [SPEAKER_02] It's just a lot happens very fast. [SPEAKER_02] It seemed to all correlate with the models hitting some threshold. And why did they, why did coding models suddenly start to click over the last couple of months? [SPEAKER_01] Like, was there a research trick? [SPEAKER_01] Was it just got enough code data in the pre-train? [SPEAKER_01] Why did it suddenly start to work? Yeah, it's a great question. We wondered a lot about why several people crossed that threshold at the same time. I'm sure it's a number of factors, but model intelligence, just the raw reasoning horsepower. I think it was like what you're doing, you know, you're doing a lot of things. [SPEAKER_02] And then, like many other endeavors, once you know something's possible, it's much easier to go do it with vigor. [SPEAKER_02] And then a really big one in the last few weeks, where it's like, OK, this is going to be the primary interface to a computer for me. And is everyone using it for coding? Or are you starting to see usage diffuse into other domains? [SPEAKER_02] I think the most adamant users are still using it for coding. But there's been just this tidal wave of people coming into Codex recently. And I'm really trying to understand what has happened that this is causing this. And the depth of what people are using it for, or starting to use it for, has surprised me. So certainly our ambition is for it not just to be about coding, but to be about all the work you do in front of a computer. And I would say we're maybe 10% of the way there for the non-coding stuff. But now that we see what's happening, now we have a real user base using it in these other ways, I think we'll get good at it very fast. [SPEAKER_01] What do you think will be the next domain that subjectively feels like it has this big unlock after coding? [SPEAKER_01] Like, is there going to be spreadsheets? [SPEAKER_01] Will it be performance reviews? [SPEAKER_01] What's it going to be? So I think there will be a lot of, first of all, I do think coding is a little bit special. [SPEAKER_02] And these models are a great fit for coding. The world needs so much more code than currently gets written. There may be no other domain that is quite like coding. But there will be a lot of others that I think are close. But the next kind of coding-like thing that I think will happen is not any specific domain, but the realization of how much time people waste trying to use a computer. [SPEAKER_02] And the idea that you can do a huge percentage of your day in a very different way. [SPEAKER_02] And maybe you don't realize how much time you spend clicking between messaging apps and copying and pasting stuff and responding to very boring things that you can clearly automate once. But the degree to which most people will realize they can sit back and watch an AI do most of their drudgery is going to surprise people. And in my own experience trying to work that way, it actually gives me much more enjoyment of work. I didn't realize how much the little stuff drags me down, gets me out of a happy flow state or whatever. So the subjective quality of life improvement is huge. Are you an OpenAI user? I am. I am. Are you OpenAI users here? [SPEAKER_01] I am. We have some good news for you all coming. I am. and watch an AI do most of their drudgery is going to surprise people. And in my own experience trying to work that way, it actually gives me much more enjoyment of work. I didn't realize how much the little stuff drags me down, gets me out of a happy flow state or whatever. So the subjective quality of life improvement is huge. Are you an OpenClaw user? [SPEAKER_02] I am. [SPEAKER_02] I am. [SPEAKER_02] I am. [SPEAKER_02] Are you OpenClaw users here? I am. [SPEAKER_02] We have some good news for you all coming. I am. You can just tell them. I am. OpenClaw has been one of my biggest magic AGI moments ever in the field. I remember the first time someone told me about it. They were trying to explain it. And I was like, okay, that all sounds cool, but I can make a lot of that work. And then it was a real reminder how when the models cross some threshold and also the product designer gets a handful of critical ideas really right, it's like a much more magical experience than it sounds like. [SPEAKER_01] I find I've been a tempting OpenClaw evangelist and trying to describe that experience you just recounted to others. [SPEAKER_01] I find it a difficult experience to communicate. [SPEAKER_01] I mean, it sounds prosaic, right? [SPEAKER_01] It's a stateful chat GPT session that can also make some use of tools and so forth. [SPEAKER_01] What do you use your OpenClaw for? [SPEAKER_01] If you scroll your message thread, what do we see? So this is a very embarrassing thing to admit. The thing I always try first... You're in the perfect place. The thing I always try first with a new kind of AI system of any sort is to... I'm a home automation nerd. And so to try to build a better home automation interface system because it's all... It never works. It never is good. And OpenClaw was the first time that I was able to get a setup that I was happy with. I also built a messaging app that I had always wanted to work. I've since switched it to something I built with Codex, but OpenClaw was the first time I was able to... [SPEAKER_02] I'm sure, like you, I could just drown in the messages and it's a very unpleasant task to wake up in the morning and have to go through all this stuff. So I was like, alright, I'm finally gonna be able to automate this. And that was... Again, should have been doable with previous systems. Hard to explain what it's like when it all actually just works and you trust that it's going to work. [SPEAKER_01] I was testing the new Link CLI that we just launched today in preparation for a launch. [SPEAKER_01] And I asked my agent to... [SPEAKER_01] You know, it means you can easily get a single use card that you can use on any business. [SPEAKER_01] And so I asked my agent to go and buy itself a gift. [SPEAKER_01] Just anything on the internet for under $20. [SPEAKER_01] And it chose to buy itself an HTTP zine from Gumroad. [SPEAKER_01] Wow. [SPEAKER_02] Yeah, there's all of the stuff that feels... [SPEAKER_02] No matter how convinced you are intellectually that this is not a real thing wanting a real gift for itself. [SPEAKER_02] And no matter how much you're convinced, okay, this is a weird emergent behavior and I'm not supposed to read into this. [SPEAKER_02] There are these things that feel strange. [SPEAKER_02] We're gonna have a party for GPT 5.5. [SPEAKER_02] And I wasn't quite sure. [SPEAKER_02] We're gonna invite people who are big users and whatever and I wasn't quite sure what to do. And so on a whim, I was like, I'm gonna ask 5.5 what it would want for a party for itself this morning. And I did. And it was a beautiful set of things including, you know, here's what I would want for the flow of the party. Here's what I would not want. You know, you should do it on May 5th. That would be funny. I would want, you know, only a short little toast and I want it not to be by me but by the people that built it. I would want a big central suggestion for 5.6 and I would want you to feed them all into me and I'll make sure we work on that. [SPEAKER_01] And now there's real moral pressure on you to... Well we're gonna do it. But it was a strange thing. [SPEAKER_01] So I want to ask you about OpenAI itself. [SPEAKER_01] You know, a lot of crazy OpenAI, I mean it's now... [SPEAKER_01] I am aware. [SPEAKER_01] It's an 11 year old organization now? [SPEAKER_01] Somehow feels like so much longer than that but yes, I get... Just over 10. [SPEAKER_01] Yeah, okay. 10 years. A long 10 years. I can't remember pre-OpenAI life that well at this point. It feels like it's been so long but yes. You've a singularity looking forwards but also a singularity looking backwards. So what's the craziest OpenAI story that's never been told? [SPEAKER_02] I mean it sounds so prosaic relative to the crazy drama that's happened but... There was this period after we had finished training GPT-4, there was an eight month period inside of OpenAI where we were all using this thing. We kind of knew that it was dramatically better and different and gonna unlock a bunch of things in the world. And no one outside the company or almost no one outside the company knew about it. And there was this... We'd walk the halls sometimes. We were like, you know, are we engaging in collective psychosis? Do we just... Have we gotten totally whipped each other into this frenzy? And there was no feedback to keep us in check or sane from the outside world. And it doesn't sound that weird relative to crazy board drama or Elon trial or something. But living through it was an unbelievably strange time. [SPEAKER_01] What's the Sam Altman management style? [SPEAKER_01] If I'm working for you directly or maybe indirectly, I'm leading some product or something. [SPEAKER_01] Or just... [SPEAKER_01] What does that look like? [SPEAKER_02] I'm definitely not a hands-on manager. I'm very much the... I'm of the style that you get great people. [SPEAKER_02] And there was no feedback to keep us in check or sane from the outside world. [SPEAKER_02] And it doesn't sound that weird relative to crazy board drama or Elon trial or something like that. [SPEAKER_02] But living through it was an unbelievably strange time. What's the Sam Altman management style? If I'm working for you directly or maybe indirectly, I'm leading some product or something. [SPEAKER_01] Or just... What does that look like? [SPEAKER_02] I'm definitely not a hands-on manager. [SPEAKER_02] I'm very much of the style that you get great people, give them a very high level thing to point at, and try to let stuff just happen. [SPEAKER_02] I think there have been two main phases of OpenAI and we're heading into a third. [SPEAKER_02] And the first one was we were only a research company. [SPEAKER_02] We were trying to figure out how we were going to build AGI at a time when it sounded completely, completely crazy. [SPEAKER_02] And we really had no idea what to do. [SPEAKER_02] And then there was a second phase where in addition to continuing to do that, we had to figure out how to build a product company. Now, we have to, in addition to both of those two things, figure out how to build this mega, mega scale token factory for the world. I think of what we're doing as building a new utility. And people are going to want to use a lot of tokens, a lot of intelligence in all sorts of ways. We need to make that as smart, as cheap, as abundant, as easy to use as possible. And that will require pretty deep full stack integration and a massive, massive infrastructure buildup. The thing that I didn't really appreciate between the phase one, phase two shift was how much my management style had to change. Running a research lab and running a product company are two extremely different things. And I suspect this third phase is going to be very different yet again. And so I've been reflecting on, if we're going to really go do this, how I'll have to change. And I think it's not going to be a natural fit for my management style. So I either have to find someone or a few people great to hire or I have to figure out how to do things in a different way. Or I have to build an AI that can manage this new thing. [SPEAKER_01] When I interviewed Jensen here two years ago, he told me and his several thousand closest friends about his 60 direct reports. Do you have any super weird, not that that—I shouldn't call his practices weird, but do you have any unusual practices like that? [SPEAKER_02] I think the closest thing that I have to anything like that is I probably talk to via Slack or text with a few hundred people at the company a day. [SPEAKER_02] Very quick, one, two messages, whatever. [SPEAKER_02] Not done by an agent—I actually do it. And the context I get from that sometimes is very helpful in these diffuse ways. [SPEAKER_01] I find there's an interesting watershed of pre-Slack organizations, post-Slack organizations, and they're truly quite different. Totally. I, like many other people, hate Slack, but I can't imagine having to still communicate via email or whatever we used to do. That's roughly where Stripe is. [SPEAKER_01] Yeah. [SPEAKER_01] So, okay, I want to talk about something you kind of just referenced. [SPEAKER_01] There is a view that the AI labs are going to progress up the stack, gobbling up the value chain voraciously. [SPEAKER_01] All these things that are certainly within the software sector, but perhaps even other sectors. [SPEAKER_01] And that there'll be this incredible positive feedback loop and runaway and a hegemonic force that we should all be getting very concerned about. [SPEAKER_01] What's your view? I think some of them do want that. We don't. One of the things that I've always admired about Stripe is it is very clear that Stripe is aligned with its customers. You know, we make more revenue. We charge our customers more. Thank you, by the way, the partnership with Stripe when ChatGPT launched was extremely critical and I don't think anyone else could have scaled that quickly. But we scaled, we pay you more money, it's very aligned and we're all happy and you just provide a layer of infrastructure for the internet. The internet gets bigger, you're happy, your user happy, it's clear what the alignment is. I don't know exactly how to do this yet, but I would like to get to a model for OpenAI that is similar. I would like us to be an infrastructure provider. I'd be happy for us to be a forever low margin business as long as we can be huge and growing fast. And I would like us to supply an intelligence meter—I don't know what quite to call it—that companies can buy, that they can use to automate things, accelerate things inside their company, they can use to build products. People can buy it, people can take it with them. [SPEAKER_02] And we find ways to really align ourselves with the success of the entire gigantic distributed economic engine of the world. [SPEAKER_02] I believe that will work. I believe the switching costs of AI—it's going to be hard to have huge margins in AI anyway. You've seen recently how easy it is, many people have seen, to switch from our competitors' coding product to ours. This is actually a consequence of AI getting smarter. It gets easier to do things like this, it gets easier to just say, "Hey agent, go do this thing for me." But if we can provide a utility and people build on top of that utility, and we think of ourselves as that kind of company, I think that can be quite powerful and very aligned. [SPEAKER_01] Well, we're happy to share lots of tips and tricks for being a low margin business. But so many people have been either implicitly critical or in certain cases explicitly critical of OpenAI for procuring so much compute. And I think, not the Codex users. [SPEAKER_01] Right, exactly. [SPEAKER_01] So, I think you were quite noteworthy for, as early as—I don't know exactly, but on the order of two or three years ago—stipulating what at the time sounded like preposterous figures with respect to the magnitude of the build out that would be required. [SPEAKER_01] And obviously, the preposterousness of those figures now looks less tenuous by the day. [SPEAKER_01] Thoughts on compute, CapEx, the buildup? [SPEAKER_02] Yeah, it's going to take a lot of money. [SPEAKER_02] I think this will clearly be, at this point, the most expensive infrastructure project that the world has ever undertaken. [SPEAKER_02] The revenue is ramping to meet it, so people feel better about that. [SPEAKER_02] Also, the efficiency gains that we've all been finding are incredible. [SPEAKER_02] So we're going to get way more out of each GPU than I thought we were going to. [SPEAKER_02] But, as has often been remarked, the demand goes up more than linearly as you drop the price of each unit of intelligence. [SPEAKER_01] And, obviously, the preposterousness of those figures now looks less tenuous by the day. [SPEAKER_01] Thoughts on compute, CapEx, the build out? Yeah, it's going to take a lot of money. I think this will be, clearly at this point, the most expensive infrastructure project that the world has ever undertaken. The revenue is ramping to meet it, so people feel better about that. Also, the efficiency gains that we've all been finding are incredible. So we're going to get way more out of each GPU than I thought we were going to. But as has often been remarked, the demand goes up more than linearly as you drop the price of each kind of unit of intelligence. Particularly if you can drop the price and the speed with which you get it back. So this question now of what is enough? I don't have a good answer to. I don't think, in some sense, I think demand for intelligence at a low enough price is effectively uncapped. Now, I was going to say we're not, but maybe we are. We're not going to build the Dyson Sphere and then just cover it with data centers. But maybe we do. Space data centers? Good luck with that. I don't even think he's that serious about it. [SPEAKER_01] I don't, myself, think that we're in a CapEx bubble. I'm not an expert in this. This is not Stripes business. But just the figures I see relative to the magnitude of the demand. It looks reasonable to me. If we were in a CapEx bubble in the future, how would we tell? [SPEAKER_02] People love to proclaim bubbles. [SPEAKER_02] And I can't articulate why, but intellectually I kind of get it. [SPEAKER_02] It does feel fun and it feels smart. Journalists in particular love to talk about bubbles. So there's ample desire to write about this when anything looks a little bit silly. And clearly sometimes it's right. There clearly are bubbles. But how you can discern between the amount of time someone calls a bubble and the amount of time you're actually in a bubble, I have never figured out how to do. In my previous career I was an investor, so I was quite interested in trying to see if I could come up with some sort of framework for this and figure out when you're supposed to deploy capital or not. And I never was able to figure it out. I went back and I read what smart people had said at different points in history and I was like, oh, they called it exactly right. But then I read a little more and they said it ten more times than the ten previous years. I don't know. I don't have an answer. [SPEAKER_01] Economists are the people who have called eight of the last three recessions. [SPEAKER_01] But they're so happy when they're right. [SPEAKER_01] So your business, OpenAI, depends in a very significant way on super talented people. [SPEAKER_01] And the difference, as I understand it, between the 20th most talented person versus the fifth most talented person versus the most talented person might be quite large and quite consequential. [SPEAKER_01] And these super talented or effective people, they're not in every case super easy to work with. And, look, some of them are wonderful people and some of them are the most fantastic collaborators and some of them are very iconoclastic and strong willed and they get easily frustrated. Just the full spectrum of the human condition. But I guess I'm curious, in a domain that's so sensitive to this efficacy and skill and talent and so forth, kind of intersected with all the foibles of humans as they exist. How do you think about this? Do you guys tolerate prima donnas? Do you tolerate them more than you used to, less than you used to? [SPEAKER_01] Do you try to manage them in a special way? [SPEAKER_01] How do you think about managing elite skill here? Someone was working on this book, OpenAI, and said to me, I think I figured out the thing that you sort of were really great at and kind of did uniquely well in making OpenAI happen. And I was like, I would love to hear. I have no idea what the next sentence is going to be. I could not predict the next sentence. And they said, you figured out how to get a lot of people who all thought they were the only capable or most capable person and everything had to go their way to work together long enough to figure out the breakthroughs. And that was the magic of OpenAI. Okay, so what's the trick? A lot of pain. I think we had very, even when people didn't like each other, and even when people thought they were much smarter than other people or had a better approach than other people, we had a few deeply shared convictions. We did kind of collectively believe in scale and concentrating resources and that we were going to do this one thing. And we thought getting this right was important enough that people were going to put aside various personal conflicts. One of the most unusual things about OpenAI was that when we trained GPT-3, the vast majority of our compute at the whole organization was going into this one single research program. [SPEAKER_02] And we would talk to people that we were trying to recruit from DeepMind at the time. [SPEAKER_02] And they would say, that's insane. It's going to create this terrible culture. [SPEAKER_00] We love music. [SPEAKER_00] We love music. [SPEAKER_02] Anyway, they would say we're talking about managing unusual personalities. [SPEAKER_02] Yeah. They would say you have to divide your compute equally, otherwise you'll have this very toxic competitive culture and all these other things. [SPEAKER_02] And we would just take the approach that we're going to embed with conviction on this. [SPEAKER_02] It's not going to feel totally equal, but this is the right thing. [SPEAKER_02] We do think we know the thing. We do think we know the direction we really want to go in. [SPEAKER_02] And they would say, well, you might be wrong. [SPEAKER_02] We have to do all those other things. You have to have this research program. [SPEAKER_02] And having a culture where we said we're going to have conviction and do this and ignore the distractions was great. [SPEAKER_02] We hope this is a memorable session for all of you. Thank you. So John and I have been doing this thing at Stripe for quite a while now. [SPEAKER_01] We started out in 2010. You and Greg started out in 2015, and to your point, you ventured through many trials and tribulations and stratospheric successes and all the rest. [SPEAKER_01] That was a nice little gloss over, but go ahead. [SPEAKER_01] Thoughts on... [SPEAKER_01] It's not easy to work successfully with a co-founder for now more than a decade. And having a culture where we said we're going to have conviction and do this and ignore the distractions was great. We hope this is a memorable session for all of you. Thank you. So John and I have been doing this thing at Stripe for quite a while now. [SPEAKER_01] We started out in 2010. You and Greg started out in 2015, and to your point, you ventured through many trials and [SPEAKER_01] tribulations and stratospheric successes and all the rest. [SPEAKER_01] That was a nice little gloss over, but go ahead. [SPEAKER_01] Thoughts on... [SPEAKER_01] It's not easy to work successfully with a co-founder for now more than a decade, and for things even after a decade to seemingly work as well as they did from the beginning. [SPEAKER_01] Just thoughts on that partnership. Why has it worked? [SPEAKER_01] And how have you guys made it such a success? [SPEAKER_01] Obviously, you and John knew each other for longer than Greg and I did. [SPEAKER_01] But Greg and I did know each other for a long time before OpenAI. [SPEAKER_01] And I think having the shared history really helps. One of the things that I had observed at Y Combinator was that one of the biggest predictors of success was had the co-founders known each other for a long time, or at least relative to their lives for a long time. And the teams that came together seven days before applying to YC on a co-founder matching site or whatever, that didn't work too often. It was not impossible. I think there were one or two cases where it did work, but it was rare. So we had known each other for a while, and we had a sense of shared values and history and ecosystem and we were clear on what we wanted to do. [SPEAKER_02] And I think we had this deep mutual respect and complementary skill set that has just worked really well. [SPEAKER_02] I'm extremely grateful. [SPEAKER_02] I think having to go through any startup experience, but particularly an intense one, without a co-founder you have a deep connection and trust too, is really hard. [SPEAKER_02] I've watched people do it, but it's very hard. [SPEAKER_02] So I am extremely grateful that we've gotten to do this together. [SPEAKER_02] Thank you. Thank you. Thank you. On an adjacent topic, we've been talking about OpenAI, but then there's this entire ecosystem of companies and startups and enterprises that are building on the platform. It's obviously an interesting moment in startups given, on the one hand, the ability to build products and generate revenue at seemingly unprecedented rates. And certainly we see this in the Stripe data, like the number of businesses reaching thresholds, meaningful thresholds is far faster than it ever has been before. You are one of the most prolific and successful startup investors ever. You, of course, ran Y Combinator. [SPEAKER_01] Have the traits that make founders successful changed in this era, or is it the same thing it's always been? [SPEAKER_01] There was a time when we used to make fun of the idea guy. [SPEAKER_01] There were these people who wanted to start a company, and they'd say I have the best idea. I'm not going to tell you what it is. I have the best idea. I just need a coder to build it for me. And then I'm going to be in great shape. And we would make fun of these people. They weren't that successful. And it was personally annoying to me, because it would be saying I have a great idea for a song, and I just need that guy with the guitar to make it for me. And so it didn't work. NYC had a version of this, which is teams without non-technical founders are difficult to get work. All of a sudden, it's the revenge of the idea guys, which is actually awesome for the world. I'm happy I'm here for it, for sure. But for a long time, I think the most important ingredient that I looked for, YC looked for, that part of our industry looked for on a founding team was technical talent. And that's still very important. But now, people who just really deeply understand their users and can't code at all, I want to fund those people. And that's a big turnaround. [SPEAKER_02] How does one think about startup investing these days? [SPEAKER_02] Because on the one hand, you have a couple of years potentially to AGI or ASI or the singularity, who knows what. [SPEAKER_01] And then you have investing time horizons or funds with 10-year time horizons. [SPEAKER_01] How does that all fit together? [SPEAKER_01] Does it? So I think to do anything at this point on a 10-year time horizon requires a real suspension of disbelief. [SPEAKER_01] And yet, that's probably the right way to live your life. [SPEAKER_02] I don't think it works to say there's this singularity in three years or five years or whatever. [SPEAKER_02] We can't see past it. [SPEAKER_02] And so we're going to do nothing. [SPEAKER_02] Or we're just going to give up. [SPEAKER_02] Or we're going to go crazy or whatever. [SPEAKER_02] You have to live as if stuff's just going to keep going in an understandable way for a long time. [SPEAKER_02] How far ahead does OpenAI plan? [SPEAKER_02] I mean, we sign 20-year power and land agreements. [SPEAKER_02] And for the product? I think we have a clear vision of what things can look like in two years. [SPEAKER_02] And then it gets much hazier after that. [SPEAKER_02] So there was in the relatively recent past a narrative that GPT wrappers and companies of that ilk were, I guess, as the pejorative suggests, undifferentiated and flimsy and be swept away by a rising tide of model improvements. Whereas now, it feels like that narrative in some sense is flipping somewhat. Instead of talking about wrappers, we talk about harnesses. [SPEAKER_01] And harnesses are seen as having this significant heft and importance. [SPEAKER_01] And I guess I'm curious for your view on this and how you view businesses for which AI is a critical enabling component and their prospective durability. [SPEAKER_01] I have had the same view all the way through, which is you as a business want to be on the side of hoping that AI gets smarter. So in the early model days, if you were the GPT wrapper and you were patching some kind of weakness in the current model that was clearly going to get better with the next model, if the next model was much better, you were sad. If you were doing something that got better, you're making any of the wonderful services that people were building with the models that benefited from intelligence, you would be happier. [SPEAKER_01] a critical enabling component and their prospective durability. [SPEAKER_01] I have had the same view all the way through, which is you as a business want to be on the side of hoping that AI gets smarter. [SPEAKER_01] And whether... So, in the early model days, if you were the GPT wrapper and you were patching some kind of weakness in the current model that was clearly going to get better with the next model, if the next model was much better, you were sad. If you were doing something that got better, if you're making any of the wonderful services that people were building with the models that benefited from intelligence, you would be happier. I think the same thing in the world of harnesses. I think the right way to think about this is data center, model, harness, that whole thing is just one cluster out of which comes this very usable intelligence. But there are so many things to go build where you're just happier for that whole cluster to get better and better and better. And then if you're secretly hoping it doesn't because you're patching some weakness in that, probably the next model crank turn somewhere in that stack is just going to solve it. When you look at the organizations that are making the most effective use of AI today, if you meet OpenAI customers constantly, large and small, if you think about the top three that have impressed you the most or the one that's impressed you the most. [SPEAKER_01] What specifically are they doing that's different? [SPEAKER_01] Everyone here knows AI is a big deal. [SPEAKER_01] We should make enthusiastic use of it, et cetera. [SPEAKER_01] But what specifically differentiates those which, in your opinion, are employing it most effectively? [SPEAKER_01] A few different directions there. [SPEAKER_01] A friend of both of ours, Toby Lütke of Shopify, was the first CEO I knew that just said we are going to be all in on AI and the way we run our company. And he got his hands dirty just building AI automation of everything and made his team do it. And he said we're just going to figure out how we take all these things that are bad and make them good with AI. And it was not a token leaderboard. [SPEAKER_02] It was not some other kind of gamified, hackable thing. [SPEAKER_02] It was just the CEO of the company said we are now going to put AI into everything we do. [SPEAKER_02] And I'm not going to be happy with you if you're not doing that. [SPEAKER_02] So that energy has now been done by other people. [SPEAKER_02] But when the CEO of a company just says we're going to automate ourselves, accelerate ourselves, however they phrase it internally, and then really holds people to it and ideally does it themselves, that has worked very well. [SPEAKER_02] So I think we're going to try this new experiment where we start sending an FDE to work with just whatever the CEO of a company needs, work with them to automate their job as much as they can. And if you just do it for the leader of a company, there's a nice fractal effect throughout the company. So that works and we'll try to help companies do that. A second thing is being uncomfortably permissive with data access. There are huge reasons not to do this. And I'm stopping short of recommendation. I'm just asking what I've seen from the most effective companies. This is easier for small startups than companies that have a lot of sensitive data and a lot of process and compliance in place. But saying we are going to record our meetings. We are going to let this AI have access to our code base. We are going to let this have access to every Slack message, every email, everything. And every employee at the company is going to get to use it that way. [SPEAKER_02] It is amazing watching these two or three person startups with AI doing everything work. And I don't know how the world is going to decide the trade-offs on data privacy versus AI efficiency. And I think there are regulations going to have to change for that. But it's so powerful. Tempo is a new blockchain that Stripe incubated with Paradigm, which obviously we're partners with you guys on. [SPEAKER_02] And the mainnet just launched, but the project launched back last summer. So it's a relatively new and small team, a couple dozen people. The Tempo team set up a harness, a tool in their Slack installation for orchestrating pretty much everything at the company. You can just ask any task, go and read these Google Docs, turn those into a bunch of linear tasks, then go write a pull request to implement them, then go deploy them and use our log analysis tools to test if the deployment actually worked. And the agent will happily go and employ tool use across all of this. And it's extremely trippy watching a small organization do everything in a single Slack channel. And I don't think that would scale to Stripe. It was the first time I had the experience you're describing, which is it's clearly incredible for them. [SPEAKER_01] I don't quite see how to transpose it for us, but this is really something. [SPEAKER_01] It's really something to watch. [SPEAKER_01] And I find that a lot of people just can't—this is where there's a big overhang—they have not yet been able to wrap their heads around the fact that you can just ask us anything and it'll probably happen. [SPEAKER_01] I myself still find myself not trusting quite enough that it's going to be possible. I don't know exactly how it's going to transpose to bigger companies. It does feel we're missing one more abstraction there, how humans and AIs are going to interface at massive scale. The advantage that these smaller companies have is the AIs. They don't have to figure out the interface with all the people. But we'll figure it out. Open source AI. Where's it going? Does it have a future? [SPEAKER_01] For sure. [SPEAKER_01] Right now, people clearly want smarter, faster, cheaper frontier intelligence. [SPEAKER_01] And most of the demand is there. But there is also a lot of demand for open source. [SPEAKER_02] And I expect that to increase relatively over time. [SPEAKER_02] So I want to talk a little bit about science. [SPEAKER_02] Because I know it's something you're very excited about. [SPEAKER_02] But it's also relevant to something I spend some of my time on, which is the Arc Institute. And OpenAI is, in fact, a foundation, a non-profit, and recently made a grant to the Arc Institute. Yes. [SPEAKER_01] And so maybe we'll get to the details of that in a second. [SPEAKER_01] But first, you want to speak a little bit to AI as applied to science, what you're seeing. [SPEAKER_01] And most of the demand is there. But there is also a lot of demand for open source. And I expect that to increase relatively over time. So I want to talk a little bit about science. Because I know it's something you're very excited about. But it's also relevant to something I spend some of my time on, which is the Arc Institute. [SPEAKER_01] And OpenAI is, in fact, a foundation, a non-profit, and recently made a grant to the Arc Institute. [SPEAKER_01] Yes. [SPEAKER_01] And so maybe we'll get to the details of that in a second. [SPEAKER_01] But first, you just want to speak a little bit to AI as applied to science, what you're seeing, [SPEAKER_01] and just how you think about grant making generally in the context of the OpenAI Foundation. [SPEAKER_01] Okay. [SPEAKER_01] So generally on the AI and science question, I hope that this will be the most important contribution of AI, [SPEAKER_01] of this technology, to human quality of life over time. [SPEAKER_01] And that if we can start to discover new science at a much faster rate, which can be new materials or cures to diseases or any number of other things, I believe that to a first order approximation, life gets better because we understand science better, and then we figure out how to build stuff with it and distribute it to people. Starting with the models of a few months ago, but really now with 5.5, the models have gotten smart enough that excellent scientists are saying, I am able to figure out better ideas, the models are able to make some small but important discoveries, and the pace of science is going to increase. Eventually, we'll have automated labs and robots and building who knows what, and we'll be able to do science much faster. But if we can start doing a decade of science, what it would have taken us in the old world in a year, the compounding effect there and what we'll be able to do and discover will just be extremely great. So I think this is going to be incredible, and this will be one of the big areas of focus of the Open AI Foundation, is money and expertise and technology to accelerate science, and trusting that that will flow to the world in wonderful ways. [SPEAKER_02] And this is going to be a big foundation. [SPEAKER_02] Yeah, this will, I think it is, it's one of the biggest, maybe it's the biggest, [SPEAKER_02] I think it will be the biggest foundation in the world. [SPEAKER_02] And yeah, so we're really focused on science and then AI resilience, [SPEAKER_02] like helping the world through this transition with this new technology in it. [SPEAKER_02] But we were thrilled to get to support ARC. [SPEAKER_02] I think it's clearly the best AI and bio effort. [SPEAKER_02] And if we can make even a small contribution to, [SPEAKER_02] with this technology and with the capital and the foundation, [SPEAKER_02] to helping make people healthier, treat diseases, [SPEAKER_02] this whole cluster of what we can do as we get better at understanding biology, [SPEAKER_02] we will be very thrilled. [SPEAKER_02] I thought that was going to take longer, and looking at the incredible work the ARC Foundation is doing, I now think it maybe won't be that far off. So in a podcast, midway through, you might hear a little interstitial ad. This is your interstitial ad for the ARC Institute. [SPEAKER_01] There is an ARC Institute booth downstairs, [SPEAKER_01] and you might wonder why there's what it's doing at the Internet Economy Conference. [SPEAKER_01] And the context here is, [SPEAKER_01] so the ARC Institute is an organization we started four years ago, and its goal is to produce hopefully the first cure for a complex disease in humans. So a complex disease is one that involves some genetic factors and some environmental factors. [SPEAKER_01] So you can think of most cancers, most autoimmune disease, [SPEAKER_01] most neurodegenerative disease, for example, [SPEAKER_01] as being a complex disease in this kind of specific sense. [SPEAKER_01] And humanity has never cured a complex disease, not one. [SPEAKER_01] We've cured lots of infectious diseases. We know how to screen for monogenic diseases for this one genetic mutation that undergirds it. We've never cured a complex condition. So we started ARC Institute. This is the goal. Alzheimer's is the first complex disease that we're targeting. And our hope is that with both new genome engineering technologies like CRISPR and then the amazing advances in AI, that we'll be able to make some hopefully meaningful progress. [SPEAKER_01] And it's only four years old, [SPEAKER_01] but the early results are very encouraging. [SPEAKER_01] The ARC Institute is currently looking for a CTO. [SPEAKER_01] We had one CTO do a sabbatical at ARC last year. [SPEAKER_01] His last year or year before? [SPEAKER_01] It was last year. [SPEAKER_01] And his name was Greg Brockman. [SPEAKER_01] And he did some great stuff. [SPEAKER_01] He helped us train EVO2, [SPEAKER_01] which is the largest biology foundation model ever trained. [SPEAKER_01] But we're looking for a full-time CTO. [SPEAKER_01] And we thought that, [SPEAKER_01] perhaps that person might be in this audience. [SPEAKER_01] And if not, their friend might be in this audience. [SPEAKER_01] So if you know somebody for whom that sounds of interest, [SPEAKER_01] go check out the ARC stand downstairs. [SPEAKER_01] And that's your interstitial ad. [SPEAKER_01] I think it's so much better that you label that [SPEAKER_01] as the interstitial ad rather than just doing it. [SPEAKER_01] That was good. Okay, well, we haven't talked about Stripe. So you were the, I think it was second investor in Stripe. Was YC the first? [SPEAKER_01] You and Paul at the same time. [SPEAKER_01] Okay. [SPEAKER_01] In the same kitchen, in fact. [SPEAKER_01] So maybe you were first, in fact. [SPEAKER_01] I don't remember in which order the checks were handed over. [SPEAKER_01] Yes, this was then his Palo Alto kitchen. [SPEAKER_01] Yeah, that's right. [SPEAKER_01] So, you know, we were two pimply teenagers [SPEAKER_01] proposing building this financial services institution. [SPEAKER_01] It sounded like a bit of a ludicrous proposition. Why did you invest? Honestly, he didn't tee me up for this. You and John were two of the most, I'd seen a lot of founders, [SPEAKER_02] and you and John were two of the most impressive founders [SPEAKER_02] of any age, [SPEAKER_02] but certainly of pimply faced teenager age [SPEAKER_02] that I had ever met. [SPEAKER_02] I clearly was right about that. [SPEAKER_02] And I had known of you, [SPEAKER_02] and I had heard Paul talk about you, [SPEAKER_02] and I'd known of you on the internet. It sounded like a bit of a ludicrous proposition. Why did you invest? Honestly, he didn't tee me up for this. You and John were two of the most, I'd seen a lot of founders, [SPEAKER_02] and you and John were two of the most impressive founders [SPEAKER_02] of any age, [SPEAKER_02] but certainly of pimply faced teenager age [SPEAKER_02] that I had ever met. [SPEAKER_02] I clearly was right about that. [SPEAKER_02] And I had known of you, [SPEAKER_02] and I had heard Paul talk about you, [SPEAKER_02] and I'd known of you on the internet. [SPEAKER_02] And I was struck that you were solving a problem for yourself [SPEAKER_02] and that it fit a trend that I thought was going to be big [SPEAKER_02] in the world. [SPEAKER_02] I really believed that commerce was going to move online [SPEAKER_02] in a huge way, [SPEAKER_02] and there were going to be lots of startups, [SPEAKER_02] and both of those suggested that this would [SPEAKER_02] could be really big. [SPEAKER_02] But I don't know. [SPEAKER_02] I was a believer that if you can find really smart, [SPEAKER_02] and still am, [SPEAKER_02] if you can find really smart founders [SPEAKER_02] and a market that's going to be big, [SPEAKER_02] you should just invest, [SPEAKER_02] and that's it. [SPEAKER_02] And based on your, [SPEAKER_02] well, just based on your general perspective in the world, [SPEAKER_02] but then also OpenAI's use of Stripe [SPEAKER_01] and what you've seen from that [SPEAKER_01] and what you see OpenAI needing [SPEAKER_01] and needing in the future and building the business and so forth, what's your advice for Stripe on navigating this, the AI era? Before that, just to check my memory, the thing you were building, the reason you realized you needed payments [SPEAKER_02] was you had built this iPhone app [SPEAKER_02] to download all of Wikipedia [SPEAKER_02] because you were going somewhere crazy offline, [SPEAKER_02] and it was hard for you to take payments for it. [SPEAKER_02] And that was, [SPEAKER_02] Yeah, good memory. [SPEAKER_02] Yeah, okay. [SPEAKER_02] I hadn't thought about that in a long time. [SPEAKER_01] I think my advice more towards [SPEAKER_01] Stripe as a company itself. [SPEAKER_01] Yeah, going forward. It's a crazy time in the world. [SPEAKER_01] There's a lot changing, a lot happening. [SPEAKER_01] And, again, you've seen Stripe [SPEAKER_01] from the perspective of a customer, [SPEAKER_01] so what are your complaints and feature requests? [SPEAKER_01] Well, it's more that I want to, [SPEAKER_01] now that I'm thinking that we should be thinking [SPEAKER_01] more like a Stripe style model, I have a bunch of questions. I probably have more to learn from you than you do from me here because I remember when investors would outsmart themselves, saying, oh, Stripe's going to be a commodity when everybody gets big, they're going to just build their own thing. And it turns out that there are multiple payments providers, and yet in practice, if you make a great product, if you make a great thing, people are still going to need to take money in the future, and they're probably just going to [SPEAKER_02] stick with you [SPEAKER_02] if you're a good, reasonable ecosystem partner [SPEAKER_02] and don't do crazy things. [SPEAKER_02] And I think that's going to keep working. [SPEAKER_02] I actually think every company [SPEAKER_02] clearly does need to get more efficient with AI, [SPEAKER_02] and they will do that. [SPEAKER_02] But this whole mindset that [SPEAKER_02] every company is going to go away [SPEAKER_02] and everything is going to be completely different, and maybe it'll be agents that need to handle payments between each other instead of consumers and merchants, but clearly money is going to have to move somehow. And my advice would be adopt AI, especially internally. Use AI to build better products, but don't assume that the [SPEAKER_02] entire socioeconomic system completely reconfigures. [SPEAKER_02] I think the world has gotten [SPEAKER_02] a little bit delusional about this. Apart from AI, [SPEAKER_02] as we look at it over the next decade, just what are the technologies and areas that excite you? Sort of AI at the kind of model and product layer. I'm obsessed now with data center infrastructure. I think there's so much cool stuff to do there. [SPEAKER_02] I think there'll be amazing new technologies at the physical layer. Energy, robots. Those are maybe that stack is probably what I think about the most. It does seem like the world is finally making a little progress on brain-machine interfaces. I'm excited about that. I am excited about... Well, I'm both afraid of and I'm excited about progress in biotech. But I'm certainly hopeful that that can get much better quickly. And I think defensive biotech is about to become very important, unfortunately. I'm excited about new kinds of computer interfaces. I think we are in an insane area right now where we're stuck with these old devices and old operating systems. And we have this magic new enabling technology. And it feels Codex is amazing, but it feels very broken to be telling this thing to use my computer. And then it's clicking around. And there's all the stuff that was made for a person, but doesn't really make sense for an AI to go use. We can do so much better there. I think there's a whole new internet protocol to make, too. So, those things. When do you think we'll have the world's first profitable nuclear fusion reactor? Depends how far electricity prices get pushed [SPEAKER_01] by data center demand. [SPEAKER_01] Maybe sooner than we thought. to be telling this thing to use my computer. And then it's clicking around. And there's all the stuff that was made for a person, but doesn't really make sense for an AI to go use. We can do so much better there. I think there's a whole new internet protocol to make, too. [SPEAKER_02] So, those things. [SPEAKER_02] When do you think we'll have [SPEAKER_02] the world's first profitable nuclear fusion reactor? Depends how far electricity prices get pushed [SPEAKER_01] by data center demand. [SPEAKER_01] Maybe sooner than we thought. I'll guess in the next five years. It's a bold prediction. I hope so. Thank you. [SPEAKER_01] Thank you. Thank you. Just going to hope we're on a roll here. Hypersonic commercial air travel. Don't follow it as closely. [SPEAKER_01] I... [SPEAKER_01] Hypersonic meaning Mach 4? [SPEAKER_01] Sure. A good chunk of time. I don't know. More than 10 years. Okay. Maybe a little less. Maybe a little less. But something like that. Are there any domains of science and technology [SPEAKER_01] that are not in the discourse in a significant way that you think will be accelerated a lot by AI [SPEAKER_01] and will have broad social consequence and impact? [SPEAKER_01] Yeah. [SPEAKER_01] The one that I think just does not get enough attention [SPEAKER_01] is material science. It's a very... It's not a cool thing. It's not a cool thing. And I think people underestimate how much of the world is materials, how much of what we depend on, [SPEAKER_02] and how much progress AI can make. [SPEAKER_02] It's such a beautifully AI-shaped problem. Just getting new catalysts. Totally. That I expect very rapid progress there, and that it'll impact all of our lives in a very positive way, [SPEAKER_02] and it gets very little attention. Last question. [SPEAKER_02] It feels that, in some sense, [SPEAKER_02] at least some version of AI is inevitable. [SPEAKER_01] There's lots of people rushing towards it, and building it out, and creating the data centers, and all the rest. [SPEAKER_01] There's this sense of determinacy. [SPEAKER_01] How do you hope that your specific involvement [SPEAKER_01] changes the trajectory for the world [SPEAKER_01] relative to some other counterfactual? [SPEAKER_01] I believe in democratization, [SPEAKER_01] and personal agency, [SPEAKER_01] and access, and that everybody deserves a really great life. [SPEAKER_02] The most controversial decision we made in the history of OpenAI was what we now call iterative deployment. [SPEAKER_02] But a lot of the thinking at the time that we released ChatGPT was, [SPEAKER_02] this was insanely dangerous to do. [SPEAKER_02] You can't do this. [SPEAKER_02] Only a small set of people who have been thinking about AI safety can know what's coming. It's an info hazard to tell the world, and it's too dangerous to ever release. We have to keep this locked up, and we will, in our ivory tower, discover these wonderful things, and we'll share the fruits with the world, but we'll have the AI and we'll control it. And that sat very poorly with me. And I thought then, and I believe now, that it is extremely important that we avoid that kind of power concentration, and that we build this for the world, and the world gets to use it in lots of ways, some of which will be good, not all of which will be good. But by enabling people to explore this very wide opportunity space in front of us, [SPEAKER_02] messy at times though it will be, [SPEAKER_02] and obviously we'll put guardrails on it for reasonable safety, [SPEAKER_02] we will give the world a gift, [SPEAKER_02] the world will build a much bigger gift on top of it for all of us. [SPEAKER_02] And if you don't enable people with this technology, [SPEAKER_02] and if you try to keep it locked up, [SPEAKER_02] which again, this may sound obvious, [SPEAKER_02] but this was the rough consensus plan of people working on it before we came along, [SPEAKER_02] I think that would have been really bad. [SPEAKER_02] I am a believer in entrepreneurship and innovation, [SPEAKER_02] and that people are mostly good and mostly do amazing things with tools. So I think my single biggest contribution has been, will be, pushing for this to be a democratized technology that people get to use and build on. Sam Altman, thank you very much. Thank you very much. Thank you. Thank you. [SPEAKER_01] Thank you. Thank you. Thank you. [SPEAKER_01] Thank you. [SPEAKER_01] Thank you. Thank you. Thank you. [SPEAKER_01] Thank you. [SPEAKER_01] Thank you. Thank you. Thank you. Thank you. Thank you. and the world gets to use it in lots of ways, some of which will be good, not all of which will be good. But that by enabling people to explore this very wide opportunity space in front of us, messy at times though it will be, and obviously we'll put guardrails on it for reasonable safety, we will give the world a gift, but the world will give, the world will build a much bigger gift on top of it for all of us. And that if you don't enable people with this technology, and if you try to keep it locked up, which again, now this may sound obvious, but this was the sort of rough consensus plan of people working on it before we came along, I think that would have been really bad. But I am a believer in entrepreneurship and innovation, and that people are mostly good and mostly do amazing things with tools. So I think my single biggest contribution has been, will be whatever, pushing for this to be a democratized technology that people get to use and build on. Sam Altman, thank you very much. Thank you very much. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.