Comment by fauigerzigerk
6 hours ago
>Companies using the value of their shares to fund demand for their services.
That's not what's happening here though. Google isn't using the value of its shares to fund demand. Google is using its own cash flow to fund this demand from Anthropic.
The question is whether Anthropic has demand from end users for the capacity they are buying from Google (that's a yes I guess) and whether that demand is profitable for Anthropic (that's a question mark).
True.
Regardless, (a) it's ability/desire to make such investments is still driven by stock-driven optimism and (b) these transactions' "signal" can have a similar, warping effect.
In this case the transaction creates demand for Google's services and also funds anthropic's growth... which represents demands for google's services.
"Loop" is an approximation of an analogy. The risk is that enough of such transactions create a dynamic that distorts feedbacks.
>(a) it's ability/desire to make such investments is still driven by stock-driven optimism
I don't think it has much to do with the stock price at all. Current platform oligopolists fear the rise of new platforms. They want a foot in the door for strategic reasons.
What could happen is that frontier labs like Anthropic and OpenAI never become platforms and turn out to be providers of a largely commoditised, low margin service.
In that event, current valuations are too high. But Anthropic's valuation doesn't seem extreme to me. Their $30bn annual run rate is valued at $380bn.
Given this price and Anthropic's strategic value, Google's investment seems reasonable.
But OpenAI/Anthropic are not selling the compute as they're buying that from Google/Amazon/etc.
So they're selling the transformation, or the model. Or the ability to make a model. And their brand and their harness.
And it seems like the model is definitely not worth 380 billion. Models depreciate incredibly fast. There are lots of models and the other models aren't that far behind.
And it seems like the harness is not worth much as there's already open source alternatives that people claim are better.
And all these companies are paying lots of money for these AI training experts.
But I suspect that any regular Hacker News reader of 10 years dev experience could become a training expert in months if allowed to play with a load of compute and a lot of data for a bit.
Just like any of us could have become a data scientist, this stuff is not particularly hard. Random horny dudes on the internet are putting out loras and quantized models in days against the open source image models.
So what's worth 380 billion exactly? The brand?
These valuations just look really off. Not by one order of magnitude, but more like by 4 orders of magnitude. Like 380 million might be a reasonable valuation, but not billion.
What I also don't get is that it's pretty obvious to me that the Europeans should all be spinning up their own, not necessarily massive, data centers and throwing a few billion at some guys in Cambridge or Stockholm or London or Berlin to make their own AI models.
Only the French have done it.
But instead the rest seem to be trying to court Anthropic or OpenAI to build data centers. Which is just stupid politics given what's happening in the world right now.
4 replies →