Prof G Markets
00:04:37 · Ed Elson
[00:04:37] S4: If it does hit $200,000,000,000, it seems like this might be a business that that is feasible. The trouble is it seems like it seems like a a seriously big if. I mean, we we had revenue of $17,000,000,000. We also know that, you know, growth is obviously slowing, but also that prices appear to be coming down. And they said this, they had these price reductions from q two to q three. They're now coming out with these discounts largely due to the open weight models that we've been seeing that have been kind of flooding the market coming from China. I mean, is 200,000,000,000... Is that is that reasonable? Should we take that seriously? What do you think?
00:05:16 · John Foley
[00:05:16] S5: I mean, as you say, there are there are plenty of risks weighing on this, and the price war is a big one. The discounts that they put on their products, it seems, were quite small according to this new amendment, maybe, like, 3% of revenue in discounts. But really... And and if they do make 200,000,000,000, by the way, then a valuation of something like 1,400,000,000,000, which is kind of roughly in the ballpark of what we're talking about, wouldn't be ridiculous compared with companies like SpaceX. If you believe this, SpaceX's valuation is same too. But my my issue with this IPO from the beginning has been that it is almost impossible to predict with any certainty where these revenue numbers are gonna go. And we also have in this latest amendment some new tweaks to the risk sec... The risk factor section of the prospectus. Not huge ones, but things that remind you that there are some sort of, you know, unpriceable factors that may weigh on this. The one that they've talked about in the latest edition is this idea that they may be getting very close to or have passed what they describe as the moment of dramatically accelerated progress attributable to AI, which is like when AI starts basically improving itself more broadly. And when that happens, then they are supposed to step on the brakes in certain ways to make sure that AI is being developed safely. What does that do to the revenue? It's... Like, it really is impossible to say at this point.
00:06:28 · Ed Elson
[00:06:28] S4: One of the more interesting developments in AI over the past week was something that came from Elon Musk. He tweeted... Someone had pointed out how Grok, which is basically SpaceX's AI chatbot, their agent, that it wasn't working. And someone tweeted about this. And then Elon responded to that tweet. I'm just quoting Elon Musk here. Quote, the cost of serving Grok was unsustainably high relative to revenue, which basically encapsulates the entire concern about all of this. I mean, about OpenAI, about SpaceX, xAI, and also about Anthropic. I mean, if Elon Musk is telling us that that is what's happening, that basically the cost of training these things and then running these things is larger than the revenue. And as a result, he's having to shut parts of it down or dramatically pull back at least on spending it seems. I mean, it makes me think that Anthropic is headed for the same scenario. And now that we've seen the $900,000,000,000 and we don't have exact clarity on how that's going to play out, but we can assume based on what we do know that it's gonna be something like a $100,000,000,000 at least per year over the next several years. That's by the way very conservative because we know that there is more than $400,000,000,000 in commitments that they're going to need to pay over the next four years. So it just makes me think that's probably where things are headed, but maybe Anthropic can turn it around. I mean, what did you make of Elon's comments? And to what extent does it say something about Anthropic's future?
00:08:11 · John Foley
[00:08:11] S5: Yeah. I like money is finite. Right? So we we are focused a lot on the spending, but someone has to pay for all of this stuff in order for it to work out. Like Anthropic has in its... Going in its favor, rapid growth in what it calls enterprise business, what everyone calls enterprise business. So selling services to companies who tend to be able to pay more than consumers, who tend to be stickier, to use the jargon, than consumers. Of course, that... The money there is limited too, but Anthropic does have a head start. Grok is very much at this point in consumer product. If you're sent to sell consumer products, how do you fund them? You can either charge a subscription a bit, or you start looking at things like advertising, which is of course how Google pays for search. But there again, like, money is finite. You know, even companies like Google and Meta are now starting to spend beyond their means to fund this AI. So if they can at least fall back on enormous advertising businesses, which Anthropic and OpenAI cannot. So at some point, yeah, the... Like the question is, do you have to reduce the level of service because you just can't afford to pay for all those tokens? We'll have to wait and see. But Anthropic is leaning hard into the enterprise space, the idea that companies will pay for its products because they deliver some kind of measurable advantage that isn't otherwise available.
00:09:27 · Ed Elson, John Foley
[00:09:27] S4: Just looking at their their compute costs, in the current quarter. So, you know, in the first quarter of twenty twenty six, it was 3.3 or almost $3,400,000,000. The next quarter, it was $5,600,000,000. This quarter, it was $9,900,000,000. So it's getting progressively larger, but also the revenue is expanding dramatically as well, which makes you think, oh, it's okay. But, I mean, it seems as though these compute commitments that are coming down the line are going to absolutely balloon the compute costs. The quarter on quarter growth rates for revenue has fallen to something like 47, 48%, in these latest q three numbers, but the quarter quarter on quarter growth in compute costs went up almost 80%. So at the [00:10:12] S5: moment, that's increasing more quickly than the revenue. [00:10:15] S4: It seemed as though when the IPO came... The... When the original filing was... Came out, you know, there was a lot of, I would say, sighs of relief at the fact that their operating losses had come down so dramatically from last year. It came down to something like $700,000,000 negative operating loss. And the previous year, was more than $2,000,000,000 in that same period. It seems as though it's going to continue to go up. I guess the question becomes, do we actually think that this company is going to become profitable in the near future? Is it actually going to be profitable in 2027? Which I think was something that a lot of people seemed to believe coming out of that first filing? I wasn't sure of it myself. But what do you think?
00:11:00 · John Foley
[00:11:00] S5: In the second quarter, they, on an operating level, had achieved a positive profit, I think, and that's now swung back to a loss. So... And and then on an accounting basis, their loss is obviously huge because they have this sort of bizarre non cash paper loss based on the value of securities and derivatives that they've issued. So like predicting when it gets profit is also again really hard. Like I sort of feel like I'm abdicating my responsibility as a financial journalist and someone who analyzes stocks by saying that you just... It's just who knows. But I really do feel with this company is you are being asked to predict things that are impossible to predict. You could argue that this company doesn't really belong in the sort of sausage casing of a traditional public company because these things are so variable. But here we are. I think ultimately, Anthropic will tell us what it thinks it's worth and and we will pay it or we will not depending on whether we think that Anthropic is gonna be one of the winners. But like... But in terms of are they gonna achieve an operating profit? Well, you'd hope so. If they're if they're cash burning, then they have to find the cash somewhere else. They have to plug that gap elsewhere. One option is by doing it through debt, but as we've heard, they may not have an investment grade credit rating, which will make that more expensive and more difficult. The other is to just keep raising more equity, maybe a bit of both. So there there is gonna be a lot of pressure on them to show that they can be self sustaining from a cash flow perspective. And at the moment, they are not.