alt Hacker NewsI don’t think the question is so much of AI is a future market, it is has the capital been allocated appropriately. And no one knows. The challenge Anthropic has is they have a narrow lead on the use case that makes a lot of money today, but new models come out every month and one tiny slip or wrong decision and they are toast. There is also a reliance on the idea that two things continue to happen:
1. Revenue growth increases 2. There are more highly profitable uses cases like code.
Neither are guaranteed.
I also think the narrative that these companies are profitable on compute is erroneous. You can’t ignore the other costs and no they don’t magically go away. Anthropic can’t stop training new models now or in the foreseeable future. Gemini 4 Carbon is about to come out and is apparently competitive. Google has other revenue sources. They are in the ball game. They can wait for Anthropic to run out of money.
Replies
Even if Anthropic and OpenAI go to zero, the bubble popping still won't hurt many people like it did in other recessions.
Actually there are two different things going on here
1. model layer companies: OpenAI, Anthropic
2. hyperscalers, data centres, hardware layer
When people say the bubble will pop, they usually mean the second one.
> I also think the narrative that these companies are profitable on compute is erroneous.
I don't think you read the Anthropic filing. It is substantially profitable if you exclude the one type of debt which was actually equity. It so happened that Amazon lent some money to Anthropic but it was tied to the valuation of Anthropic, which became a "problem" because Anthropic's valuation grew 10x. You wouldn't call this a loss in any other circumstance.
The amount of profit they make on compute is also not particularly attractive (at least right now). It ‘s only slightly more than 2x for Anthropic, far off from many big cloud services which sell compute at something closer to 20x.