The Information's TITV
00:18:55 · Akash Pasricha, Stephanie Palazzolo
[00:18:55] S0: So, like, are the model companies completely cooked then? Like, what's, what's, what are you predicting here? [00:19:02] S3: Well, it's not clear. I would say this is definitely not a death sentence for them. I think there has been this theory throughout the AI boom that actually as models get cheaper, people want to use more of them. Because the models are getting better, people are wanting to use more of them. Actually, maybe overall spending will increase over time. But I will say though that, you know, a lot of companies that I talk to, they are just, just very nervous about, you know, every year they basically have to commit to spending a certain amount with OpenAI and Anthropic. So even if they know that their overall AI spending is going to increase over the next year, they don't know exactly that it's going to increase more on OpenAI or more on Anthropic specifically, or whether that new spending will come from open source or from a third model provider, for instance. So it's just been very difficult, I think, these customers to be able to commit right now to spending, you know, whatever 25 or 50% more on OpenAI for instance, over the next year. That's just been a really tough uphill battle, I think, for the salespeople at the model developers to have to go up against. [00:20:14] S0: Now, on the other hand, you did say that companies operating routers like OpenRouter, they've had a very easy job of getting people to renew their contract sounds like.
00:20:14 · Akash Pasricha, Stephanie Palazzolo
[00:20:14] S0: Now, on the other hand, you did say that companies operating routers like OpenRouter, they've had a very easy job of getting people to renew their contract sounds like. [00:20:26] S3: Yeah. So we've really seen revenue at some of these other companies like OpenRouter explode over the past couple months, or even companies like Snowflake, where essentially they kind of act as like a central hub for a bunch of different models. So, you know, even with like Amazon Web Services, for instance, they have a product where you can go to that and then, spend across different models from both OpenAI and Anthropic. And so like what these kind of platforms offer is, you know, more flexibility and the ability to switch between different model providers very easily. And so this might push more customers towards these kind of like independent, like third party platforms versus buying directly from a model developer where they need to commit upfront to spending, you know, X, Y, Z amount over the next year on their models only.
00:20:26 · Stephanie Palazzolo, Akash Pasricha
[00:20:26] S3: Yeah. So we've really seen revenue at some of these other companies like OpenRouter explode over the past couple months, or even companies like Snowflake, where essentially they kind of act as like a central hub for a bunch of different models. So, you know, even with like Amazon Web Services, for instance, they have a product where you can go to that and then, spend across different models from both OpenAI and Anthropic. And so like what these kind of platforms offer is, you know, more flexibility and the ability to switch between different model providers very easily. And so this might push more customers towards these kind of like independent, like third party platforms versus buying directly from a model developer where they need to commit upfront to spending, you know, X, Y, Z amount over the next year on their models only. [00:21:17] S0: Right. Or they move very much into the application layer, which we've talked about you and I on the show before. I mean, as the models get commoditized, the stickiness might come from the user experience of an application or being able to do things for people, not just do the back end of things, which could very much be where this story goes. I do want to ask you, so we had some news last week. OpenAI is reported to have decreased the size of their funding round to $20,000,000,000 from $30,000,000,000 This is the private funding round that they wanted to raise. Is the story here, Steph, just that they can't raise the money? Do we know anything about the state of their business? What do you think this is?
00:22:09 · Stephanie Palazzolo, Akash Pasricha
[00:22:09] S3: Yeah. So it is unclear what is kind of going on behind the scenes. I will say that not being able to raise the full amount that you went out to raise is never really a good sign. You know, like I think in an ideal world, in an ideal world, you would want the round to be oversubscribed so that you have too many investors actually like trying to get in. I do think like a major dynamic that's going on here is that, you know, Anthropic is, about to go public. It's, you know, releasing its financials in its S1. It's putting its numbers out there. And to be honest, those numbers are really, really good, not just revenue growth, but also margins as well. I think a lot of people were very pleasantly surprised at the Anthropic numbers from last week and just how efficient they have been able to be with running and training these driving models. And, you know, we, we don't know the exact figures for OpenAI, but if they are not as good as Anthropic's, even if they are, you know, objectively pretty, pretty good, that's definitely gonna put a bit of a damper on the round and make it a very easy comparison for people to look at OpenAI's numbers and then look at these literally public figures from Anthropic and kind of do the apples to apples comparison. [00:23:20] S0: Right. And they'd get liquidity faster too because you could buy some Anthropic and it'll go public. Whatever it is, six month delay, even lockup will expire a lot quicker. In a world where everyone's talking about bubbles, it's easier to get your money out faster.
00:23:20 · Akash Pasricha, Stephanie Palazzolo
[00:23:20] S0: Right. And they'd get liquidity faster too because you could buy some Anthropic and it'll go public. Whatever it is, six month delay, even lockup will expire a lot quicker. In a world where everyone's talking about bubbles, it's easier to get your money out faster. [00:23:37] S3: Right now is a very high risk time to invest in OpenAI, right? This is probably going to be the last round before it goes public. So you have kind of limited upside because you don't really expect there to be like a massive valuation jump between, you know, the pre IPO round and the company going public. If Anthropic's numbers are better, then there's a good chance that the OpenAI IPO may not be this huge blockbuster event that will make people tons and tons of money and a huge return on their investment if they got in at whatever, you know, 1,200,000,000,000 price or whatever the most recent round is. So I think it's, it's a very high risk situation. And I think a lot of investors are just trying to play it safe right now and maybe not invest in this round. [00:24:25] S0: Right. Speaking of Anthropic stuff, so we got Anthropic's amended S1 out this morning. Did you get a chance to look at it? What should we be paying attention to here?
00:24:38 · Stephanie Palazzolo
[00:24:38] S3: Yeah. So, I mean, I think there's two things that jumped out, one of which we have already written about, but that is that the revenue growth from Q2 to Q3, although impressive, it has begun to slow a little bit. And so I think that's an interesting signal for this company that has obviously had insane revenue growth over the last year that maybe things are starting to taper out. Maybe it's facing more competition from OpenAI models or for people looking for cheaper models. But I think the other big update was that Anthropic disclosed its future, compute commitments, which were almost, you know, almost $900,000,000,000 obviously a huge number. And half of that is going to come after, you know, either in 2031 or after. So not only have they committed to a huge dollar amount of compute, but a lot of that is going to be coming, you know, five plus years in the future. And so that really does, I think show just, I don't know, I think it might lead some investors to be a little worried just in terms of like, we don't know necessarily where AI is going to be in five, you know, five, six, seven, eight, nine, ten years. And the fact that they've committed to so much compute so far out is a little worrying, especially because of this dynamic, right, where Anthropic's revenue, it's not these five year commitments that customers are signing. It's usage based. And, you know, we were just talking about how Anthropic and OpenAI are having trouble getting customers to commit to, you know, bigger year long commitments. So you have this dynamic where revenue and compute, which is their largest expense are not really, operating on the same kind of like patterns, if that makes sense.