The Exchanges

Q all time, still needed to organize his thoughts to somebody else. You are going through, I can't think, you're, you're, you're, you have almost like a singular lived experience, especially for somebody as young as you are. So, I'm curious, like, who plays this role in your life? Like, who do you go to that can even remotely empathize with what the hell you're dealing with on a day-to-day basis?

A Kind of three categories here. One, a lot of the researchers that have been here forever, we've kind of, like, all been through it together, and we've developed this set of, like, shared language, intuition, standards, whatever you want to call it, and That I have not been able to replicate with anybody outside of the company, um, when it comes to, like, the shape of what's happening and what might happen next and where the technology is likely to go. In, in terms of questions of just, like, you know, business and the world, for a long time in my career, Paul Graham and Peter Thiel have been two of the people that I have learned the most from about lots of different phases of my career and are still the two people that I go to if I really have, like, A very non-obvious problem that I'm stuck on. Uh, and there, I have not found anyone else after a lot of looking that has the same kind of ability to, to just, to like think in a super non-linear way. Like, you know, if what LLMs do are predicting what word comes next, those are two of the people that I can predict the least what word is going to come next. And that is a super valuable skill. You go with like a, oh man, I feel really stuck, and I've kind of thought through all these options, and someone that can tell you like, eh, I think none of those options are good. Here's this thing that now seems totally obvious and correct t…

AI assessment note: “Paul Graham and Peter Thiel have been two of the people that I have learned the most from”

Q You got to say more about non-standard. Can you be more specific? Are they spiky?

A You don't want to fund a founder who has A very slightly different take. On the same idea as the last thousand people you talked to has tried to convince you and maybe convince themselves that somehow they're completely different and doing something totally new, but it's mostly like trying to, you know, fit in with the herd and be on the same track as everybody else and do what they're supposed to do, which is start a startup. And, you know, they've heard Peter Thiel say enough times that, you know, there's like something that you're supposed to be doing different that they kind of try to emulate that, but they don't really mean it. Like it's, it's very Clear to me when you have someone who just thinks differently than most other people and has, is, is willing to stand by convictions that are very unpopular, may well be wrong, but if right, like, at least they're going to be really right, versus someone who is like, uh, You know, thin veneer on the same idea that everybody else has. In late, late, when we were starting OpenAI, there were very few AGI efforts in the world. There was DeepMind, one or two others that I can think of. It was like a very non-consensus thing to do. In that same year, there were probably, I'll pick on it just because it, it came to mind, but there are other categories too. There were probably many, many thousands of founders starting photo sharing apps…

AI assessment note: “someone who just thinks differently than most other people and has, is, is willing”

Q How did you work your way through that? What was your thinking? Like, how'd you do this?

A We asked a bunch of people who had been at great research labs of the past. And it had been, you know, OpenAI started as sort of a time when everybody in Silicon Valley, as their vanity project, including me, wanted to start a research lab. And there were all these books about the heyday of Bell Labs or Zioxx Park that were very popular. Everybody was talking about this. There was a huge amount of discussion. In fact, I even see one of the books over there about, ah, about Bell Labs. But there was not a ton of people that had, like, in living memory how to actually do it. So we talked a lot to Alan Kay, we talked to a handful of other people, and we got some advice from them about, you know, what made a really good research, research lab, and some of it was really good, some of it didn't translate as well to the current moment.

AI assessment note: “We asked a bunch of people who had been at great research labs”

Q people, I heard you say something that was interesting. I think ties to what you just said about kind of like we're inventing technology, the technology is increasing rapidly, but the adoption should be slow and deliberate. And I heard you on another podcast saying, Hey, I'm even considering like, How much should I let AI see every single thing that's on my computer? You want to talk about that?

A With the latest generation of models, I don't want to say they feel smart enough, because I think we should always aspire for them to get smarter, but they're pretty smart. And I feel more limited at this point by the amount of useful context the AI has on me. Like, I want the AI to know as much as it can to help me. I want it to be doing things I can't or don't want to do on my own. Like, I I'm not going to read every post on our internal Slack. I'm not going to go read every story a customer has to tell about where ChadGPT worked for them or failed them. I can't. And then there's like other stuff of like, I just, you know, I probably could read more research papers than I do, but like, oh, it's like, takes a lot of mental energy and, you know, but I would love to have an AI agent that is constantly trying to be helpful to me and that can Look at and understand more context than I can, or that I have time for energy for to do on my own and can help bring that context to bear and give me good advice when I have to make a decision. So I think we've focused Correctly. So much on model intelligence that on the product side, we have not yet thought enough about what it means to give a model more context than any person could have and help advise that person on, on their big decisions. My sense is we are just on the precipice of being able to see a very different way of working with…

AI assessment note: “I feel more limited at this point by the amount of useful context the AI has”

Q Why? Because so many people, there's a billion people using your products on a weekly basis?

A And each time we get a new level of model, we put it out in the world. And we see what, you know, what works, what doesn't work, where people need us to relax the guardrails because they have good things they want to use it for, where we have alignment failures, where we have safety systems failures. ChatGPT has only been out, like, less than four years. A billion people use it. And sensitive, important stuff. And the fact that we can deliver something that is broadly considered safe, of course there's issues with it. Like, in that short of a time frame with such a powerful technology, I think there is no way we could have done that in Ivory Tower. And You know, this is how I believe you build good, safe, robust, useful technology and products, and I think is a great, a great learning of Y Combinator. And it would have seemed to most of the AI safety people, you know, totally impossible to get to this stage and still have the level of safety guarantees we have. Now, I do think it gets harder from here, but I don't think you're going to solve it by disconnecting yourself from reality.

AI assessment note: “there is no way we could have done that in Ivory Tower.”

Q think to talk to you about this. We're going to look back on 2026 is a year that every business was up for grabs. He said somebody was going to build the AI native version of Shopify. And he said, and it's going to be me. And so at night, he is literally trying to rebuild. If, if you started from scratch, what would you do with deter current technology?

A That was the other thing I was going to say about him is he does it himself. Like he is using these tools himself. He's writing software himself. He's trying the models himself. He is trying to like re-imagine his workflow flows himself. Most CEOs, when you get to that level have like teams of people that Are managing teams of people that are trying to implement the thing, and they're trying to, like, make you happy, and they're trying to, like, you know, smooth the rough edges. I think it's very hard to get the feel if you're not actually doing the thing, and he, he does it so hands-on all night long as far as I can tell. I'm not sure if twenty-twenty-six will be the year that every business feels up for grabs. I might disagree with him a little bit there, uh, but, but I, I, I get the spirit of that, and I do understand that it feels like that's happening.

AI assessment note: “That was the other thing I was going to say about him is he does”