Uncapped with Jack Altman
00:35:16 · Jack Altman, Ev Randle
[00:35:16] S0: And then Anthropic's like, hey. By the way, BTW. We got some drugs coming. You're like, Okay, well, don't feel so good now. So I think there are some reasons that people won't want to do that. And then I think there's situations where you can post train on open source, and that can be useful. And if you've got a lot of data, that data can outperform the extra intelligence that you might get from the next model. And so you'll see some of that, too. [00:35:38] S1: I feel like we hear that pretty consistently from founders that aren't even selling an AI product but just have a big product corpus, and they're using AI throughout their products. Where they're like, look, if you really spend the resources to fine tune and you have really good trace data from the task that you're trying to do, yes, maybe Vanilla, Kimi k three, or whatever isn't quite as good as whatever the model, but you can actually get a really small model, like a really, really small open source model. And if you fine tune the hell out of it, gets to a k three or a Sonnet level performance, and it is dirt cheap. It is dirt cheap. Think we had a large public CEO come in and talk to us about that. It's like there's clearly all these use cases where that's true.
00:35:38 · Ev Randle, Jack Altman
[00:35:38] S1: I feel like we hear that pretty consistently from founders that aren't even selling an AI product but just have a big product corpus, and they're using AI throughout their products. Where they're like, look, if you really spend the resources to fine tune and you have really good trace data from the task that you're trying to do, yes, maybe Vanilla, Kimi k three, or whatever isn't quite as good as whatever the model, but you can actually get a really small model, like a really, really small open source model. And if you fine tune the hell out of it, gets to a k three or a Sonnet level performance, and it is dirt cheap. It is dirt cheap. Think we had a large public CEO come in and talk to us about that. It's like there's clearly all these use cases where that's true. [00:36:23] S0: Yeah, but then it's like the nice thing is we know the answer. The market's telling us the answer. And the answer is that the inference companies in combination are doing mid single digits worth of total revenue. That's a reasonable proxy for open source because that's where it's running. And that's whatever percent of the labs companies. And you can argue for that there'll be some mix shift. But what we're basically seeing is that open source might five, it might be 12. It'll be some percentage like that unless there's some change of the overall spend, and that's what will be there.
00:36:23 · Jack Altman, Ev Randle
[00:36:23] S0: Yeah, but then it's like the nice thing is we know the answer. The market's telling us the answer. And the answer is that the inference companies in combination are doing mid single digits worth of total revenue. That's a reasonable proxy for open source because that's where it's running. And that's whatever percent of the labs companies. And you can argue for that there'll be some mix shift. But what we're basically seeing is that open source might five, it might be 12. It'll be some percentage like that unless there's some change of the overall spend, and that's what will be there. [00:36:57] S1: Yeah, so I think the synthesis between these two is that again, shout out Eric Vitrea. It all works. It all works. But also, the way it all works is, actually, they're probably not even rivalrous to each other. And I don't know if this would be the case, but also, I'm sure the Frontier Labs, they're probably even happy about there being an open source ecosystem because it probably creates all... Like, from an antitrust perspective [00:37:21] S0: 100%. [00:37:22] S1: If it's like, oh, they're getting too powerful, it's like, no, there's a healthy open source ecosystem where people can also use. And I think that's probably also healthy. Have no idea if that's an actual thing that they think about. But right now, this will be excluding the Yeah. [00:37:35] S0: I'm not sure the scalars. But... Yeah.
00:37:35 · Jack Altman, Ev Randle
[00:37:35] S0: I'm not sure the scalars. But... Yeah. [00:37:37] S1: Yeah. But if you look at... So, like, right now, if you look at the... So we'll exclude the hyperscalars because we have no idea how much open source inference they're doing. But if you look at just the inference clouds, as we've talked about, and then you look at just OpenAI and Anthropic this is very imperfect, but it gives us some Okay proxy. By the end of this year, I think the top three or four inference clouds will probably do like $6,000,000,000 of run rate ish. And then let's say OpenAI and Anthropic will do like $170,000,000,000 of run rate ish if their growth extrapolates from public disclosures. So that's like three ish percent thus far between the inference clouds and the top two frontier labs. If the growth continues [00:38:18] S0: Yeah, and maybe if you add the clouds in, maybe it's double at most or something like that. [00:38:22] S1: That's right. So we can say 5% to 10% in terms of revenue for open source versus closed. But if you have the growth that the labs have had and that people expect again, this isn't a guarantee, so this isn't a projection but if you do have the growth for three to four years on the frontier side, and the labs combined are at a trillion dollars and open source takes 10% of that, that is like $100,000,000,000 of revenue. And the inference market has consolidated. There's really only three players. [00:38:49] S0: We
00:38:49 · Jack Altman, Ev Randle
[00:38:49] S0: We [00:38:50] S1: have Fireworks, we have Baseten, we have Modal. There's other ones that are good and all that stuff, they really are the three champions of the market, and I think we're going to see more consolidation to those three than less. So you have probably two or three companies at that time splitting $100,000,000,000 of revenue, which is incredible. Which is incredible. Mean, it [00:39:09] S0: means one of those companies could be worth $100,000,000,000 or more. [00:39:11] S1: Probably both. You have a $40,000,000,000 company. Even if it's 50% gross margin, it's probably a $200,000,000,000 company. You have hundreds of billions of enterprise value for open source, and then you have the trillions in Frontier Labs. And everyone is very happy with that. That's humongous companies all around, and it's through the synthesis of open source will always, on a revenue basis, be some percentage of frontier, and whatever that percentage is, there's so much room for the inference platforms to grow.
00:39:11 · Ev Randle, Jack Altman
[00:39:11] S1: Probably both. You have a $40,000,000,000 company. Even if it's 50% gross margin, it's probably a $200,000,000,000 company. You have hundreds of billions of enterprise value for open source, and then you have the trillions in Frontier Labs. And everyone is very happy with that. That's humongous companies all around, and it's through the synthesis of open source will always, on a revenue basis, be some percentage of frontier, and whatever that percentage is, there's so much room for the inference platforms to grow. [00:39:41] S0: By the way, and then we can move on from this, but a couple of years ago, I feel like there was a worry of, man, it seems like AI is either going to wake up the Terminator Sam god, and everything's going to be really bad. It's too good. Or it's just not going to matter that much. Crypto all over again, head fake. And at the moment and this is why I'm happy to see the market reacting positively to Anthropic. It's like at the moment, unless RSI takes us into a weird place, we might actually be splitting the middle landing, where the labs are extremely important, but they are not $60,000,000,000,000 companies, where it is not useless technology that doesn't help anybody. That's a nice thing. Seems like the happy path. It seems like the... I feel like my odd... My P happy path is higher than it used to be, and that's awesome. Amazing. Which is amazing. [00:40:39] S1: That's amazing. It's amazing. Do you wanna talk about bio? Obviously, you've you've got... I I feel like you've been really interested