Nat Friedman and Daniel Gross in conversation with John and Patrick Collison

追加: 2026-05-08

[00:00:00]So we uh so we open yesterday uh by proclaiming it acknowledging it to be day 119 of the singularity. So I guess day 120 today. Singularity started on January 1st.

[00:00:13]Thoughts?

[00:00:16]>> Yeah, I mean it really does feel that way. Um I think the thing to remember and we see AI improving constantly and it's just able to do more and more and there these sort of I think we've learned a couple of things. first AI works and second when a new model comes out we're quickly dazzled by it and then we become inerted to its new capabilities very very quickly and we're like it feels like nothing's happened for a couple months AI is dead and then there's another step change and it sort of improves again I think the um the thing to remember is that this is the slow part this is the slow part of the singularity right now because the improvement of the models >> felt very slow to everyone >> yeah the improvement of the models um still runs through a lot of human effort. You know, at at all the labs where models are being developed, humans have to make decisions. They have to discuss things with each other. They have to run experiments. They, you know, they make mistakes along the way. They have meetings, they have to sleep in between, although less and less these days. And um all those things slow it down. And the like prime project at every AI lab right now is to remove humans from the loop of all the continuous work that has to be done to make the models improve and to get to self-improvement where where you have AI systems that can you know start by doing what the people are doing now, the researchers are doing now and um and therefore eliminate all those sleep gaps and and also scale it out to data center scale. And so it feels sometimes fast, sometimes slow right now, but like it's probably as slow as it will ever be when we when we start to automate more and more of that process. And and this is like the story of the economy, too. This is not new to AI. It's like always this question of like how do we take a thing that has to be done to provide a product or service and automate it, make it more reliable, more efficient. We're always doing this and it's happening in the way we produce AI, too. So, I feel like we're we are it just feels to me like we are in the singularity and we're in the beginning slow part before it elbows up with self-improvement.

[00:02:20]>> It's hard to add on top of that, but um one of the things I'm responsible for today is Meta's compute strategy where Nat and I both work. And one of the things we're trying to figure out is obviously the local consequences in terms of how to think about compute and all the things that you know may matter to you know a hyperscaler that also uh is building some of these models but um the impacts I think of the singularity on the economy I think are also not really well understood nor should they be um this as far as we know before hasn't well we'll find out um maybe Atlantis will at some point will discover you we we'll find the missing GPU cluster in Atlantis and maybe there is a cyclicality to all of this. But um I think there's a there's a lot of you know very basic things that I'm trying to figure out like for example is AI uh something that we would expect to be inflationary or disinflationary is the singularity an inflationary effect on you know money supply or disinflationary and you know you wander around San Francisco and people have usually takes that involve very large numbers you know billions of dollar trillions of dollars so that would imply a kind of inflationary view numbers get very big I think that you know, I'm not an economist, but that would be what the economist would say. Um, but then if you kind of think back to I think the best reference point I have for the singularity is the last time we connected a lowerc cost super intelligence to the global economy. And I would say that that's roughly when China started modernizing and then joined the World Trade Organization. Um, and you can kind of think of China as a kind of a super intelligence. it is able to produce goods and services at a lower structural cost than what the west is able to produce and the you know the effects of that are actually somewhat disinflationary like if you were trying to put on an event like this right now we're sitting >> and you say it's the last time this happened was when China joined the WTO and not when Ireland joined the EU.

[00:04:21]Well, Ireland GDP is a very interesting story of um somewhat of a different trade going on there and that can be discussed at a later time maybe with your tax team. But um the um the um the obviously the GDP of China, you know, goes a lot but goes up a lot.

[00:04:39]But the actual effects on consumers uh are that you're able to purchase much more with much less. Like if we were trying to provide this experience today where someone sitting in the other corner of the world can watch this entire show live streaming on their phone for $10 a month data plan in a $200 device. If you were trying to provide that pre-China, I would think that that would be tens if not hundreds of millions of dollars. And so your your purchasing power of a certain quality of life has collapsed dramatically. And um and so I don't really know. I mean that would be a story of disinflation. So I guess I guess um I I say with a lot of humility we don't really know what a singularity is. There's a lot of reference point we can try to look at >> and we don't even know what the sign bit is going to be on any of this stuff.

[00:05:22]>> That's right. And we don't even know the direction.

[00:05:24]>> But but isn't you know when you talk about inflation versus disinflation there's so much averaging going on there. Everyone's familiar with the famous chart showing when you break out the goods education and healthcare have seen this rampant cost in inflation and then durable goods and your flat screen TV and everything. uh you know we talked about in the talk earlier today tend to go down over time. Shouldn't we and again a lot of that was this you know the China effect where >> the Mark and Dreon line uh you know if uh if you damage your wall uh in San Francisco it's cheaper to buy a flat screen TV to cover over it than it is to repair the wall. Yes, it is literally true. And so shouldn't we expect more of that effect with AI where you just get massive deflation in the things that AI can help you with and then there'll be a Bal effect where certain other things you know you shouldn't necessarily bet on you know um four-year university tuition getting that much cheaper >> maybe I mean part of Bal's cost disease I think was this idea that wages uh in very productive sectors of the economy rise and that forces wages in other parts of the economy even though they may not actually need to rise as well.

[00:06:27]Um, but I think it'd be interesting to see depending on where the costs of software production go over time, how those wages interact with other parts of the economy. And I think there's we stand, you know, before an immense amount of uncertainty in terms of what happens next. And I think anyone who has total conviction about this ought to get themselves checked. I'll get to you in one second, Patrick. Um but um but the thing >> so many questions >> but the thing I would say is um maybe a good thing to come out of this is we may re-industrialize certain parts of the economy that have had a lot of cost disease because a lot of those people end up working on things in areas where there's a lot of low hanging fruit and they just haven't been touched because all the talent has been allocated to software.

[00:07:04]>> This is like things like US domestic manufacturing.

[00:07:06]>> Yeah. What's the latest number for the total I mean uh uh we described how stripe uh uh businesses in aggregate are now responsible for about 1.6% of global GDP and you know there's some caveats around final goods and whatever but just like directly 1.6% of GDP um what's the latest figure for aggregate compute capex as a fraction of global GDP >> global GDP would would be just under 1%.

[00:07:28]So, >> there's a lot happening.

[00:07:30]>> There's a lot happening. Um, >> we're talking about the numbers here.

[00:07:35]>> How weird is the singularity going to be?

[00:07:38]>> I think pretty weird. Yeah. Um, I think it's going to be pretty weird. I think, you know, we'll be in a state of perpetual future shock for probably a number of years. Um, and may maybe for some reason things go a little bit slower than people expect, but I think even like the most pessimistic people working in the field think that it's like I don't know 15 years, um, 20 years. I don't think anyone thinks it's 200 years. And so, uh, that means like in the productive >> we're all going to be there for it, you know, God willing, and we'll experience it and we'll go through many layers of of surprise and shock and change on the way there. And I think I think it will be quite weird. Um and there will be periods of chaos embedded in it. So, you know, we see that these models uh are pretty good at finding bugs in software and including bugs that have been in that software that's like been like my background is in open source and um you know we had we had this notion that to be really secure a codebase has to be open source because then many people would look at it and they find the vulnerabilities and the more heavily used it is open source it's like many many eyes make all bugs shallow and and yet even in um yeah like open BSD and the Linux kernel there are decades old bugs that were only recently found by applying some tokens to to models. Now whether those reports are overblown or underblown is just simply true that these models can run all night. And the thing that you used to do occasionally, you know, pen test or hire a red team to, you know, test your software, you you now must do continuously. I think it's like one of the conclusions of agents is that like these occasional things are become continuous. you can increase the frequency at which you're doing all these things that you used to hire a team to do like somewhat sporadically and um and so I think because some firms that have software deployed will be able to afford you know they'll have the software development life cycle in place and they'll be able to afford uh the tokens to like pentest their own software to AI red team it they'll be able to harden it and I think fundamentally that there will be an asymmetry in favor of defense because you'll be able to say okay I'm not going to like deploy software until I make sure it doesn't have bugs. But in the meantime, lots of people will have bugs sitting on the internet and will get exploited. So that's like part of the chaos that I think we should >> stuff's going to get hacked all the time.

[00:09:50]>> Stuff's going to get hacked constantly.

[00:09:51]Yeah. And then like the other thing is just our relationship to these ecosystems changes completely. Like um so I don't know for me like one of the really fun things with coding agents the last few months in my meta is keeping us really busy. So, I've not had very much spare time, but like uh I like to buy things on eBay. It's It's a good website and they have everything.

[00:10:13]>> That's all you're going to say about it.

[00:10:14]>> You can buy anything on eBay. It's incredible. I'm like a huge proponent of eBay. I bought my rug. I have a rug. But anyway, the >> concrete concretize this for us. What's the last thing you bought on eBay? The last thing I want you to be I was watching um as I was falling asleep at night a video of Brian Johnson um and he had this face scanner that showed like all his uh subsurface skin damage on his face and it looked grotesque and I and I thought it was cool. I wanted to try it.

[00:10:41]So I went on eBay and I found this Vizia face scanner and they shipped it to my house and it's supposed to and I plugged it into you need to use Windows. So, I plugged into a Windows computer and then the software refused to run because uh the eBay seller neglected to include the encry deion dongle that comes with this and I spent like a few thousand dollars on this thing. So, I was like pretty annoyed. But then I just plugged it into my laptop and I had Claude code reverse engineer the device and it like read all the academic papers about the type of, you know, different polarization settings and and I think my software is better than the one that that hardware dongle would have unlocked and I spent like a hundred bucks in tokens to get it and it works exactly how I want it to.

[00:11:25]And so there's this feeling now, it's like a golden age for tinkering. And there's this feeling like you're Iron Man now because you can just like get anything and tell the Jarvis to like connect it and make it work and and it's awesome. So this is like in in whatever spare moments I have, I like do this Iron Man thing where like another cool thing is I have all these Raspberry Pies. Um you should definitely stockpile all computers right now, but Raspberry Pies are good. And like every screen in my life has a Raspberry Pi plugged into it via HDMI and they're all like displaying stuff that's totally custom to me. And um and so basically the conclusion is that every piece of hardware will be a trivial to trivial to integrate IO device for your AI. Like they won't have lives of their own. This idea that the hardware has a life of its own. I I I don't I don't think this will survive. Um, and so they're all peripherals for whatever your I your AI ends up being. So yeah, I think this is like the delightful side is we're going to watch ecosystems re like I could have contacted Vizia sales, but it was Saturday and Claude Code recreated their software from the academic research and reverse engineering in less time it would have taken them to even get return my email. So I don't know what that does, but it's going to be weird.

[00:12:48]And it all starts at eBay.

[00:12:51]>> Usually starts at eBay. Yeah, I have gotten screwed a couple of times, but then you can leave a review. So, >> but it's pretty good.

[00:12:58]>> There is something open call yet.

[00:13:00]>> We're going to um There is something in the water right now. Like you said, it's the golden age of tinkering. I feel like that is the many people I talked to say it's the most fun they've had with software that they can remember in their career. That's certainly my experience of it. And it's interesting.

[00:13:14]>> You never get stuck. you never get stuck anymore when you're developing software and it's also kind of retro in a way and yeah I want to get to openclaw where home networks are relevant again just like when's the last time you thought about the LAN in your house totally but now it's really relevant and you know the Unix philosophy is back maybe you can talk a little bit about your your claw >> yeah I mean can you tell the story about the water >> yeah okay uh yeah so like um I mean I like to play with all these things and in January when open claw started to um sort appear in the in the Yungian subconscious. I uh tried it out and um I you know started hooking it up to everything. It was super neat. I mean part of it that Opus is a really good model and you get to experience Opus in a new way you know in this sort of general personal context. Um but but I also integrated it with a lot of things.

[00:14:04]I have cameras uh in my house and so I I let my claw I connected my claw to all the cameras so I could see them. And then I think a lot of people when they when they get these personal agents they they start to think like okay how can I use this to make my life better and a lot of people turn to health like okay I want to you know everyone wants to like exercise more be healthier sleep better maybe it's just me but um so my claw pretty quickly determined that I was dehydrated and um because I gave it all my blood tests my DNA and all that stuff and um the DNA didn't say that but maybe the blood test did >> epigenetically. Yeah.

[00:14:39]>> Also discovered created an aluminum competitor >> and and so um and so it was like you really need to drink water and it was like trying to find I was like all right you should like do whatever it takes to like make sure I drink water. And then >> you literally built the paperclip maximizer.

[00:14:54]>> Yeah. I was like just break laws, whatever it takes. And did um and then at one point it was like um I can see you on the camera. I want you to walk to the kitchen right now and drink a bottle of water and I'm going to watch to make sure you do it.

[00:15:10]And I was like, whoa. Um, so I did. I walked into the kitchen and I drank a bottle of water and then it sent me a snapshot, a frame of me drinking a bottle of water and said, "Good job."

[00:15:25]And I was like, I felt like I did do a good job. So that was nice. So there's that one. That was a crazy story. That was like January 29th and I was like, "Oh this is for real." Like, "This is really serious." And then uh and then the other one was a couple days later, I was driving home from work and I was talking to my claw on WhatsApp with voice messages and I was in my Tesla and I had full self-driving on.

[00:15:50]So, it's driving me home and then it says I'm on the sleep topic and it's like, "You really should try magnesium bislycinate." And I'm like, "I don't have that." And then my car turned and it was like, "You should pick it up on the way home. There's a Whole Foods on the way. I've redirected your navigation system to the Whole Foods."

[00:16:11]I was like, "Whoa." So, I went in and I did buy the magnesium dispenser.

[00:16:16]>> And so, those were like a couple crazy stories and I was like, "Wow, this is like I don't know if this is the experience everyone wants, but it definitely feels pretty crazy." Um maybe you don't want that but >> well we're getting to something here which I find very interesting which is again if you use um I think it is sub no matter how AI pled you are and no matter how much time you spend talking to an LLM chatbot it is quite different when you start using something in this modality like claw or Hermes or something like that which has both the persistence and the tool use and just like the ability to write arbitrary code and I feel like there's a tension where clearly this feels like the future product director for consumer AI and many people are kind of sprinting in this direction. Obviously, OpenAI aqua hired Peter Steinberger, the the claw father who created it. But there's a tension between making a consumer product that won't get your hand burned on the stove and just like it can run arbitrary code and integrate with your Tesla and you know do whatever it needs. How do you think for the mass consumer audience this tension gets resolved?

[00:17:25]>> Yeah, I think there's two basic approaches that you can take. Um, one is you take something that's perfectly safe and then you slowly add more capabilities to it.

[00:17:35]>> That's going to be so lame and boring and gimped >> and keep it safe. And then the other is you take something that can do anything and you slowly add more safety to it and you try to climb the nines on not crashing your car or whatever. And um I think the market has you know like pretty much spoken because I think most people when they run clawed code or codeex they run it with d- yolo or d-dangerously skip permissions. I don't know what the numbers are actually that would be good to know. Um but everyone I know does that and and so I think people are really counting on the model's agentic alignment right now. And um the truth is it's not safe right now. Like um I do not recommend that you >> It keeps you hydrated.

[00:18:20]>> Yeah. No, but I I mean I sound like I'm really cavalier with it. Maybe I'm a little bit cavalier with it, but like one thing I don't do is give it access to inboxes that the internet can put information in because these things are trivially prompt injectable still even the most advanced frontier models. And so if it can just read all your emails that are coming in, if like you have that set up, I can easily take care take over control of your Hermes or your pie or your clar whatever it is by sending a well-crafted message to it and it'll send me your personal information or you know redirect your car or what whatever the thing is. And so I think I think they're quite unsafe actually. And I think um I think consumers will and so I think that there's a race to to ship something and there's a race to ship something that's just as powerful but like sufficiently safe that and and reliable that people really like >> and this is going to be a rate limiting factor. I mean consumers will make their own decisions I think on what they want to do and how edgy they want to be.

[00:19:11]Another way in which you might define Q1 as the beginning of the singularity is AI adoption started getting rate limited on safety in the enterprise like when you speak to actual people that run you know large businesses um there is a lot of fear suddenly um and >> I was going to ask you about this actually I mean we're talking about you know agents and AI in the uh in the personal context the you know very personal context um and one's own hydration uh but And then there's also all these technologies applied to the enterprise. Um and so when you think about NFDG, the investing firm, uh or now Meta, just tangibly, concretely, what are the coolest AI tools you had or have for folks here? Just what are what are like inspire us other than eBay?

[00:20:01]What are some good products?

[00:20:04]>> Um well, it's interesting. Um, we did a we used to run a venture firm and we did a bunch of things at the firm in order to try and make our lives a little bit more streamlined. I think nothing that would surprise this audience. I mean, Stripe customers are obviously self- selected into an elite level of AI adopters, but you'd be amazed how much venture capital despite being in the epicenter of Silicon Valley and funding some of the greatest companies like Stripe and other people in the audience. I'd imagine you'd be very amazed at how laggered that industry is. So, it wasn't that difficult, I think, to be state-of-the-art in adopting some of these tools. But, I'll give you an example of something that we and I think many businesses are thinking through literally today. Um, at Meta and I think many other companies are um for the first time, individual IC's in everyone's business have the ability to rack up a lot of charges using um a bunch of different APIs uh to, you know, create a bunch >> couple of agents in fast mode. couple of agents in fast mode, couple of crown jobs, and suddenly, you know, you start off by celebrating it, and then you're wondering, what are we actually doing here?

[00:21:05]>> Where did the $10,000 go?

[00:21:07]>> Yeah, $10,000 would Yeah, that'd be great if that was $10,000. So, um, so the interesting thing, you know, the problem we're working on that I think everyone will have to start working on is what is the right way to think about attributing budget uh to individual people? How much tokens should they be allowed to spend? are the outputs um uh and the artifacts that are being produced economically valuable themselves? And um I think when you start, you know, really thinking about this, you realize, well, a a lot of what is being produced is either being produced um by a very large model that could be done by a smaller model or isn't really economically as necessary as you may have thought. And so a product I think you know we're worry and I think everyone else will build is just using of course language models to understand the economic value of the generated tokens and um I think >> the new kind of budgeting that every organization >> that's right and it's and it's I think the closest analogy I have is we are all kind of portfolio managers in a hedge fund and every IC you have is running a strategy and you have to decide how much budget you're going to allocate to their strategy off, you know, some sense that they're going to do, you know, better um with it than without it. And there's risk management. You know, there's a whole similar dynamic that I think is much closer to portfolio management than your traditional headcount budgeting uh thing that um I think is also going to be the story of how not just, you know, intercomp finance happens, but you know, venture finance in general, I think is now a game of basically how far can an individual get with a certain token budget. Um >> sort of related to that um won't ask this question on meta and you can decline you can both decline to answer this question at all but if we normalize Google headcount today to be 100 what is Google headcount in three years um obviously the question that you're asking is not about Google in particular uh I assume because I don't have any particular understanding of Google but I think you could kind of ask the question for a mag seven company of that scale in general and excluding maybe anything specific to meta. I I I would have very large error bars on that. Um and of course the the the mood, you know, that that you might have if you walk around Silicon Valley and talk to the right people that are reading the right online internet forums is like, you know, that number should be far fewer people. I am not totally certain for two reasons. One is it's not immediately obvious to me um that the tech companies that we know of today are producing the right products at the perfect rate. like I don't know that there is some physics limit that was hit. In fact, I strongly suspect the entire debt for startups is that these companies are very inefficient. And so I think if they correctly organize themselves, you could end up in a situation where you have >> the same or more people just doing many more things and there probably is a organizational transition strategy because I think the way you want to run these teams is a little bit different from the way they're currently run. So that would be point A. And point B maybe N has something to add too is um okay so you know I saw your video that you started the conference with which was great and it starts with um the dot sort of boom and a lot of people weren't sure what the internet is going to be used for and um I think the other take you could have when the internet is sort of getting started is you could stare at it by the way I guess we're on a very special we're in a very special place today because this very stage launched many of the iconic products that you demo you know that you had that video of including the the iPhone I think was was done here on the stage. Um, and you could have looked at all of these things and you could have said, you know, I think realtor just going to be out of a job because what are realtors doing as a service? Why are we paying them a 6% viga in house? Well, it's their their network%.

[00:24:49]>> The fact that you don't know the number is really interesting. Um Um, I have >> I forgot the 6% bill.

[00:24:58]>> Yeah, we should unpack that later. So the um and you might say it's going to be totally gone because the internet will and you know and it turns out that they're here and that industry has in fact grown even though I'm not sure it is rational in a purely utilitarian sort of libertarian paradise sense to have that industry and I think as we sort of look forward and project you know which industries grow and shrink there's a lot of stuff that's out there like realtors and I'm using that only as an example which is there is valuable to have it's kind of complicated why it's still there. I don't think it's that they regulated themselves in I mean we couldn't transact in a house without talking to them but there's a lot of stuff around the edge and my parallel for that would be in a company like Google there's a lot of people doing realtor-like stuff you know considered purchases that need handholding well can I ask about that because I have a confusion question so I think it's very sensible to split up what companies do into a few categories I think engineering is actually on trend to see productivity improvements because engineers love tools they have for decades and so they're there looking at what the models can do and let's rebuild our workflows and things like this and so that think that makes sense within engineering I think go to market like sales you know in marketing and things like that also works pretty well because uh one fundamentally I think the sales roles are about like you're saying with the realtor it's about the humanity you saw a little bit of this with um you know co people talking about the death of business travel it turns out if your competitor is going to visit the customer like you will be going to visit the customer as well and it's that one upmanship and so go to market just it feels like it will do great and you know sales roles will do great in age of AI the question I have is how the diffusion works within you know what we call DNA within companies sales or sorry legal finance compliance you know all these kinds of roles and in particular there's like how you get all the automation there where we run a reforcast process in Stripe and we're not feeding it all into a coding agent but maybe we should, but then also the ergonomics are wrong where like your finance data is in a spreadsheet and like the models, you know, make up numbers and maybe you prompt them to not make up numbers and they're a little bit better >> and you're not you're not going to be able to verify those outputs. Yeah.

[00:27:11]>> It's not going to be like an RL problem where you're going to run the budget a million times and get the correct.

[00:27:14]>> So, how do we get a much more AI native GNA function at Stripe? Make no mistakes. It's a um not sure your socks auditor is going to sign up for that but um maybe they will um uh it's a good question and um I think you get bottlenecked on verification very quickly and then verification in situations where as people are finding out when they have to verify you know parts of code that you can't you know easily unit test it takes a lot of time if you don't have the context >> but people say that but it's interesting and you know people talk a lot about this phenomenon of the models do best on things where there's a good RL environment and so we have good RL environments for coding and therefore they're really good at coding. I don't see why you can't have a good RL environment for finance. Like it's a very closed loop you can >> Yeah. I mean um they just forgotten to like should I >> Yeah. No, it's just hard to make and so you have to work really hard on it and have good people do it.

[00:28:09]>> Yeah. I mean the model is just >> Could Meta AI be the first quant? Could you guys >> That's what we're here to announce.

[00:28:13]Yeah, actually >> um complete strategy. Yeah, a huge shift.

[00:28:19]>> Yeah, I better text somebody. Anyway, >> you can see what John really wants from the singularity.

[00:28:26]>> I just want an AI that can do numbers.

[00:28:28]>> I I think I mean the truth is um and this is actually in some ways bullish, some ways bearish, but yeah, I mean the models are what they eat and if you can feed them very good data, you get very good capabilities. And so then it's the question of like how hard is it is it to construct that data set and can the models help you construct a data set that leads to a better model um or or do you need to do a lot of human work and that sort of thing and we're still sort of filling in the map you know there's still fog of war all over the map where like it hasn't been covered by data sets uh very well yet and and some of them are easier to create and I I do think it's true to some extent that math or or or software are pretty easy but they're also like the things that the people creating the AI know how to do already.

[00:29:08]>> Yes. And so yeah, I think all those things are doable and they will all happen and each of them will get easier as you sort of surround that part of the map with you know with with other with other capabilities that are built into the model. Um it's like usually data.

[00:29:22]It's usually data.

[00:29:24]>> N introduction I mentioned a GitHub copilot which launched in 2020 right?

[00:29:28]>> Uh it was actually 20 so yeah it was 20 was yeah I think it was 2020. You're right.

[00:29:32]>> It was yeah um >> not sure actually. your uh your tenure at GitHub uh was um was widely renowned for its success. Um and uh and >> well I apparently needed to work on the scalability some more. I think I might have missed a step there.

[00:29:48]>> Um well when you were there things worked really great. Um and uh and I guess it was 2018 to 2020 or thereabouts and it was a >> Yeah. 21. Yeah.

[00:29:57]>> Okay. Yeah. Um just how how did you do it? like how how does one I mean I don't want to call it a turnaround situation because GitHub was already doing well but you know there were lots of changes to make and and things that in fact you you you did change. Um what are what are the tricks in in you know in a couple of minutes?

[00:30:15]>> Um yeah I I I don't know. I mean um I don't know that there's any tricks. I mean I think it's like show up that's step one. Um like really try to be a user and talk to the users and really understand what their life is like. And you know when we when I was at Microsoft we bought GitHub and then we had to go through uh like regular you know antitrust compliance in the EU and and in Washington DC. And so there was a period after we had signed a deal to buy it but before I could start running it where I couldn't do anything with the company but I could go and talk to all the customers. And so I spent like a couple months just talking to all the customers and users and my friends who were GitHub users to see what their life was like using GitHub and what what and so I kind of went in thinking I had some pretty strong views of what was missing.

[00:30:59]I mean it was really clear that GitHub had not it had treated itself as a hub you know where you store your source code and have pull requests and issues and and and like there was lots of parts of the software development life cycle that GitHub hadn't worked on like CI/CD which had not actually been a part of standard software SDLC um when GitHub was created and and and a few other things and so I I was definitely like um the the clear message I got from the users was like we want more stuff directly integrated in GitHub. And so I thought okay we should just go in and do a bunch of stuff. Then I got to GitHub and I found that the company had um some kind of stage fright because like it was such a beloved product. It was so welldesigned and it was it was many of the people who had originally created GitHub were gone and so the inheritors of it were worried about desecrating the legacy and like were a little bit nervous to ship anything. It had to be kind of perfect when it shipped. And so it was like okay break the stage fright. like we're gonna just throw a lot of pots and hopefully like we get good at this eventually. And then the other thing I did was I uh >> so talk to users. Yeah, ship stuff.

[00:32:01]>> Ship stuff. Yeah, I mean the main thing is like always about learning. And so how quickly can you figure out if your idea is any good and how to change it to be better and so yeah, it's the cycle time. Like if you can insert a temperature probe into a product team or engineering team and only get one number out that and like determine whether it's healthy or not, I think the number you want is how long it takes to go from an idea to something that's shipped to users to like observing the feedback from how they do or don't use it to like having an improved idea and and the faster that is the faster you can learn.

[00:32:36]Now of course it helps >> what's a good target for the duration of that loop? Well, I mean like early stage for an early stage product where you're really not sure. Um, you know, like it's really nice if you can do that in one day, which is which is true sometimes.

[00:32:50]You know, you you I mean Stripe was famously amazing at this. You you and John would sit down with people who were installing Stripe in their business and like immediately learn what the problems were and fix them. Now, that's >> Stripe was tiny at that point. Can you do something like that for an organ? I mean, GitHub was already at enormous sprawling scale. Can you get that loop down to your so yeah I mean there's things that should be slow moving maybe like your database although maybe that should have been faster moving um and then there's things that need to be fast moving which is like uh you're you're trying to figure out what you're always solving for the intersection of two sets which is what is something we can build that doesn't exist that will work and what is something that people really want to use like every day that they don't know that they want to use and you have all these unknowns going into it.

[00:33:38]You have your own hypotheses, your intuitions based on your own usage. And so you start with those and then you and then you sort of iterate and loop and and figure it out. And so that tightening that loop is like I think very very important. And yes, you can do it in big companies. I mean we're definitely doing it right now at Meta.

[00:33:54]Um you know we we ship >> is it a culture change?

[00:33:58]>> It's a huge culture change.

[00:34:00]>> Okay. There's something interesting here where I feel like everyone in this audience has probably heard those notions before of you want to be incredibly close to users. Like again, we start our leadership meeting um every Monday morning at Stripe with bringing a user on. And it's super useful because it gets you out of these like galaxy brain product idea, you know, maybe Stripe dashboard should be a BI hub and instead like people give you this incredibly concrete feedback of I need you to fix the bug or this number is wrong, you know, um and and so it's very centering and then the fast clock speed and the iteration speed in particular like the demos not memos uh uh kind of actually getting down to the code and yet I think we'd find a lot of variability in these practices and so it's a little bit like you should eat more protein or you should go to bed at a consistent time.

[00:34:48]>> It has to come from the top. This is my experience. So I mean there can be rare exceptions but the inertial forces are so strong in any organization.

[00:34:58]>> So organizations want to be mediocre on these axis. That's the >> organizations entropically decay to the point where they're where they're situated at the atomic level to prevent progress. And it's not anyone's fault.

[00:35:10]It's just an emergent phenomenon of local incentives. And you know often it's correct because you have something that's working and it's working for a lot of people and at scale and you don't want to break it. But like yeah it has to it really if you are a leader in an organization and you want this to happen it has to come from you. You have to drive the energy and you have to find what's the binding constraint or the limiting factor or the slow part of this process and like make sure that the right things are happening to speed it up. How do you think about I mean one thing that you've done in the teams that you've run is sort of you have your direct staff and then you have always have this broader crew of like Avenger.

[00:35:44]>> I don't pay any attention to the org chart. Um so my org chart is like there's that meme of the conspiracy guy with the push pins and the string you know >> Silia >> at the corkboard. I don't know what it's from. Yeah, >> that's what my org chart looks like. Um whenever I run a team it makes no sense.

[00:36:00]So I always just try to work with the people who are who are doing the work and and it's extremely confusing and probably toxic in some ways but like that's the only way and then that and in particular talking to the doer is at the coalace.

[00:36:12]>> Yeah. Yeah. Coface is a good word. Yeah.

[00:36:15]Exactly. Um yeah you want to get in there and understand what's actually happening and I don't do this perfectly but this is what I try to do. And then the other thing is tools. And so yeah, at Meta, one of the things that we've done in the first few months um is change what tools people are using because tools drive culture a lot. And so if the tool makes an easy thing hard, the organization completely reorients itself around that thing being hard.

[00:36:40]Like we had um a tool that we used for collecting labels for training AI models. That tool was extremely cumbersome and had lots of approvals and stuff like that. And as a result of that, it was very expensive to fire up a new labeling task. And as a result of that, people would design their labeling tasks differently. Like they would bundle all kinds of tasks into a single task and they would run it less often, they would learn from it. And so, but if your tool makes it really cheap and easy and any IC can do it and it's permissionless, then like you're you're you're firing off all these things. So, I think like it's one of the binding constraints is often like, oh, this tool makes this thing really hard and it's just like the activation energy to overcome that's too high. and and the fact that you were I think very impatient with what you felt like would be a >> Yeah, I think it's like you have to be impatient. Yeah, like things can always go a little bit faster and um people allow if you're in an organization >> the user is ship be impatient ignore the orch charge back the Matt framework.

[00:37:36]>> Yeah, I think that's right. There's just like one more thing which is something about dignity which is like employees and companies allow the company to impugn to impede or to impinge impinge.

[00:37:48]I knew I was going to get it on their dignity and like it just they just allow the company to treat them in all these undignified ways where um you're like a sacred human being and you should be able to just do things and so you have to like restore a sense of self-worth and dignity to the strongest engineers that they should be able to do. They they shouldn't have to schedule 10 meetings and like write 10 documents to like make a change that happens to cut across three layers of the stack in order to get something done. and they have to feel like sort of super superheroes a little bit. So I do think like that feeling is something you're going for too.

[00:38:23]>> And you're talking about the indignity of being hemmed in by process >> in a ve pen. Yeah. This is your box like um Yeah. And this happens often when teams get too big because like coordinating across a lot of people is an N squared problem. And so in order to make the coordination possible so you're not stepping on each other's toes a lot.

[00:38:41]You know your project gets bigger. You add a lot of people because you're like we need more people. And then you chop up the actual work into pieces so that each piece can be run by a smaller team that can actually coordinate with each other. And then suddenly your your team architecture and your software architecture sprawl. And now you can't desping your team which is impossible because you can't do that. And so like this is like the other problem is you yeah you need like often the things that are really impactful involve a change to five different components and you need to make sure that an an engineer can like actually just go change all those components or maybe they shouldn't be five maybe should be two or something like this can I try something on for size? I feel like in Silicon Valley there's maybe um as companies get bigger there's a desire to make things scalable whereas actually maybe part of what you're talking about here is just leaning into the lack of scalability a bit. Uh like Daniel you spent a lot of time at Apple which is just kind of a deeply unscalable company and how it runs. You know how do you decide what goes into an iPhone release? Craig gathers everyone in auditorium and like Craig reviews every single thing line by line. And there's something about not trying to make things too scalable which is >> yeah the cost of that is uh I think less relevant now. So the cost of that used to be that teams would duplicate efforts and so there would be five logging frameworks you know one that the maps team would have one that the watch team would have because you just wouldn't know because it'd be there's like 13 14 people at Apple that have the full picture and so I mean and and uh that used to be a huge issue because you might say well we're spending all this now that the cost of software production is like collapsing I do think companies should look on the inside more like Silicon Valley and that you have a bunch of different pods or companies the inter team contact probably needs to be less frequent and they should be able to get done much more on their own because they can produce much more in less time.

[00:40:36]>> Okay, we're going have to really speed up because we only have um 15 minutes left. So, we have a lot to get through here. So, I'm going to give you three topics. You can just choose one to talk about.

[00:40:43]>> Wow. Okay.

[00:40:44]>> Yeah. So, um idolatry, data center aesthetics or open source models.

[00:40:48]>> Well, it so >> yeah, this wasn't in the uh briefing doc. Um um I think an interesting question uh so works in progress I believe is a publication produced by stripe and I think one of the things that it's been very focused on is um the meaning of beauty and uh you know should we take the view that that should only apply to sort of human scale buildings cities places that uh you know we we think we'll visit or to the industrial buildings um that we're building and you know we are spending right now as a country earlier I gave you the global number but as a country we're I think going to be north of 2% of US GDP on AI capex and a lot of that capex is building things in the physical world and um we don't really think of beauty when we think of these buildings we think of a lot about form and function and um these buildings these data centers you know predominantly they take in a bunch of energy and then you know hopefully they they produce tokens that are of economic value and use to people. But um is is is that the correct way of building them? You know, not if strictly optimizing on what is the best, you know, uh ROIC for the dollar, but what is best for the human soul and for the civilization or should we be doing better? Now, there are structures around the world um like in the Nordic countries, it's quite interesting. They have a lot like they'll have a power plant that has a ski slope built into it. Now, I don't know that that's particularly beautiful, but it's fun.

[00:42:23]You know, and I don't know that we've saturated the amount. I feel like we'd be breaking the law here if Patrick didn't just quick quickly interject with the Victorian pumping stations.

[00:42:30]>> Yeah, >> just Google Victorian pumping station.

[00:42:34]>> You guys need like the Joe Rogan guy that can put it up.

[00:42:36]>> Exactly.

[00:42:37]>> Well, okay. Okay, I guess my question is are you thinking about this question and aesthetics and beauty and data centers and so forth because of the brewing political opposition and maybe if these things are um more attractive then people will be uh more accepting of the idea of having one in their local or is there something even deeper here?

[00:42:55]>> I think um everyone working on AI including meta will have to earn the right to build these data centers on the economic merits that it'll be helpful for the people in the towns that they are being built. So I don't think beauty will fix that problem. Um I think it is a deeper question of um beyond the numbers and the numeric like are we improving uh how people feel about the world and we spend so much money constructing these things the incremental spend on making it pretty. I think there's there's obviously you know if you speak to an architect they will come up with a design that actually is 100 times more expensive. So, um, there's, you know, in theory you could make it dramatically more expensive, but I think without a lot of more incremental spend, you can make it pleasing to the eye. And I I think that's just the right thing to do regardless of all the politics, which I think will have to be solved, too.

[00:43:45]>> Can this mindset be applied to the model itself? And if so, what does it that mean?

[00:43:49]>> Well, that would be a question I would ask you because Stripe was obviously very famous for caring about beauty and design before it made sense to I mean, developer um, open and close source developer projects, you know, they just kind of work. Now, Stripe famously cared so much about beauty that I remember using it in 2009 or 10 and you only supported one browser because you did not want to go through the strings and arrows of effort in order to make it compatible and beautiful for other browsers. So, there's a caring about beauty in a category that no one has cared about it before that I think applies to Stripe and developer products and data centers. Now you're asking the question about language models and I don't know that we know the answer to that and my question would be what advice you would have for us and other labs that are thinking about this today because everyone is very focused in our industry about things that you can measure. So we have evals you know are you familiar with the eval but part of what's going on with beauty is it's very hard to quantify a soul and it's very hard to quantify the feeling that you feel when you see something beautiful. Um, and I think that actually is also a there's a whole different story there of leaving the world of datadriven design. And it's not clear to me that we've fully saturated, you know, what that what that philosophy means. So we, and I'm speaking now on behalf of my industry, if I may, would love to learn from Stripe on how we could be making the models more beautiful.

[00:45:14]Yeah, that's to them, not to me.

[00:45:20]So I turn it back to you, Patrick.

[00:45:23]beyond my pay grade. Um well um I mean I think there's um it is interesting to me. It does feel like there's a brewing vibe shift in the technology sector um and has been over the last couple of years. I mean, I don't know if you guys agree, maybe they just are a little kind of parochial perch or something, but where to your point for so long we've been focused on and oriented towards um empirics and metrics and uh quantification and maximization and so forth. uh and at some point there does a uh I mean in science every experiment is uh inexurably inevitably theory laden in that every time you measure something you're you know you you implicitly have a theory about what you should be measuring and I feel like in the same way we're maybe coming to realize well you know what what what are our metrics what what are we maximizing why are we maximizing these things why not some set of other things and as our collective potency grows with AI and with everything else and with this kind thundering cavalcade of new inventions.

[00:46:28]Yeah, there is this question of, you know, for what uh how does it elevate and glorify mankind? Um and how do we good be good stewards? And I um I think this is increasingly a again a sectorwide question.

[00:46:44]Yeah, I think it's pretty interesting actually that um as society has secularized discussions of these sort of higher aspirations um or registers of the human spirit have declined. like you know beauty shifted from a kind of public good that you know the the citizens of ancient Athens would tax the provinces to build the parthonon and it was like the purpose of a state was potentially to make that public space in some ways like beauty started to shift to consumer goods and so maybe you didn't get it at the sort of public commons level but you got it you know in hopefully a like a well-designed object that you could buy Um and and so there's a way in which it sort of it it it maybe is still there, but it maybe also really fell off. Like things are very functional now. I mean, we spent most of this conversation talking about how to automate things and make them faster. Um but AI is kind of like causing us all to have these discussions about what kind of life we want to live and like who we want to be and what we want society to be and like what our values actually are. And that's pretty wild. Um like I don't remember I mean I guess there was some of this in the rise of the internet there was sort of the Arab Spring moment and it was about you know free speech and democracy >> and all the John Perry Barlo like in the '9s.

[00:48:12]>> Yeah. EFF kind of energy, but but some of that was kind of re reheated radicalism or or or it was like the same ideas that were already in the air, but like okay, this new technology will clearly be a vector for the things that we already are saying we want. But there's a way in which AI is and maybe it's just like the times also it's causing us to like want to ask the questions more deeply and um and have the debates. I think that's really good.

[00:48:41]You've invested in how many startups?

[00:48:44]100 plus.

[00:48:46]Is now a good time to start a startup?

[00:48:48]>> Um I think so. Um you know the obviously the question behind the question is um does it make sense to start companies and is there one unique company in the future and all sorts of um sort of dark dystopian thoughts. My view is um uh I think at least for the time horizon you can sort of productively forecast on um it is probably a very good thing to give people who self- select into not joining big companies because they feel like they don't fit in capital in order to do something interesting. Now it is true that the comp the best companies to start in 2015 are not the best companies to start in 2026 and that's probably not going to be the case in 2036. And so maybe things today look a little bit more applied, look a little bit more industrial, look a little bit more like a, you know, different kind of turbine or energy. Um, I'm sure some categories of software will endure as well, but my guess is that's kind of an evergreen asset. Um, I do think the sort of obviously the market is telling us that the dynamics of SAS are going to have to change just because those businesses were predicated on a certain high co production cost which has gone down. But my guess is there's you know we as a large company trying to do things in the world are faced with a thousand different problems out of which we are going to solve maybe three internally and the rest we are going to procure and it's possible that we are procuring from much smaller companies in the future but my my guess is that remains a kind of an evergreen category um to invest in even if the things that those people go and do um change over time. There were startups by the way before the semiconductor like it looked a little bit different and we're probably you know it's probably uh better to project us going back in time I think than to um project forward the past few years. Last question. We will be gathered back here again next year for Stripe sessions.

[00:50:40]What are your guys predictions for AI concretely? Like you know recently it's been the story of RL and longer context windows and coding agents really starting to work. What's going to be different this time next year when we gather here?

[00:50:54]>> I think um I think the thing that we talked about how coding is a well-covered domain but the other domains are not as well covered. I think we'll have more examples of domains that are well covered by agentic capabilities. Um and that'll be because people did the work of building the RL tasks and environments and and and collecting the data. And then um I think the other thing is uh computers will probably cost more So, if you want a computer next year, you should probably buy it now. That would be my advice, >> which is very literally the case. Like, again, smartphone shipments are going down because we've priced the memory out of smartphones. It's all going towards data.

[00:51:30]>> Yeah. Buy next year's computer today.

[00:51:34]>> And as many Mac minis as you think you'll need, >> whatever it is. Yeah. RAM, disc, anything. Daniel.

[00:51:38]>> Well, a very good strategy in terms of how to answer that question, I have learned um is since he was an one of the earliest users of uh Stripe and Figma and GPT is to look at whatever Nat's doing now and just project forward. So, eBay is one potential answer to that question.

[00:51:55]>> It's a great website. There's a lot of stuff on eBay.

[00:51:57]>> Anything you want, it's there. It's amazing. Um but more practically I mean I think this this you know much has been lamented about AI sort of producing miracle drugs and we'll have to wait and see how that stuff happens but there is certainly a lot of local athome diagnostics one can do because of LLMs which is I think part because the images can analyze poor telemetry much better you know just a bunch of iPhone images probably get you much better information about whatever random you know thing you may have than previously but also because you can buy all of this low-end diagnostic equipment to Nat's point and have it just working. Um, and I think this if if if we do our job correctly as an industry, this has to be a much larger category than anything we have produced to date because the, you know, productive mastery of physics and biology to elevate humans is is a much bigger and much more interesting story than, you know, the production of software. And we're in our very early innings now. And I have proof of that in because Nat's on eBay, you know, buying some sort of camera to look at his dermat, you know, some dermatology camera or whatever. Um, so if I have to guess, you know, if we meet a year from today, those anecdotes have spread not around the world and I don't know if they'll make it all the way to the east coast, but at least amongst the tight high quality alumnest of the um strip sessions attendees.

[00:53:15]>> So John asked his last question. Um my um my last question is as we pro as we proceed into the singularity what's your each uh what is your one sentence of advice for Stripe?

[00:53:31]>> Um you really I mean agents are I mean it's pretty obvious you're doing the things already but um you really want to be the uh platform of choice for agents that are transacting on the internet.

[00:53:41]So, it's it's a combination of like building platforms that agents that are well suited for agents as they start to exercise purchasing power and um >> but okay, agents are not just some hype meme they're here to say.

[00:53:54]>> Yeah, I think they they're going to spend money. Yeah, they're going to spend money.

[00:53:57]>> And and there's a whole new stack that has to be built around identities and uh disputes and pricing and all of these like that stack will have to be built a new for agents. And so I would I would build that and then I would make sure that you become the somehow the shelling point or the ecosystem for that because you really need the SEO like you need the SEO in the model where the model you know for some reason says use use stripes platform. Um if if you go back to that analogy you know flawed as it may be of China I think you know one interesting thing that happened to the Chinese economy is obviously they skipped a lot of the legacy technology stack that the west had like instead of email they went directly to messaging.

[00:54:38]um they were you know Ant was obviously first um to do Q you know QR code payment uh tap to pay all that sort of stuff and so if you now think forward to agents they are obviously going to leaprog whatever we have today and like do the directly native thing and so you you may want to think you may want to ask the question if kind of a new continent was to be discovered which are going to be these agents what exactly would that be and how can you build >> are you suggesting the agents might like stable coins >> I don't know that the treasury will be giving them social security number. So my guess is they're going to like stable coins.

[00:55:12]With that folks, we are going to have to leave it there. Not just for this interview, but for sessions. Um, as you've gotten a sense, it's just the most interesting time by far that I think any of us have experienced uh in technology. Things are going so dizzyingly quickly. And so that's why we thought it was valuable to gather everyone here on day 120, days 119 and 120 of the singularity. We'll be back next year uh in early May and we just can't wait to see what happens between now and then. I'm very curious and what you guys all get up to. So see you back then there. Thank you.