[danielciocirlan]

Is China’s strategy sustainable, given the enormous costs of training frontier models?

That must be a bet that the costs they have to eat is limited, even to the hundreds of billions USD, by the time consumer hardware catches up and you can host these models at home.

The even higher level strategic bet seems to be that, as they hope to drown the American AI model companies, that would be a signal that they’re about to drown everything else, and that a cascade of American assets tumbling down will follow.

[znnajdla]

> given the enormous costs

The costs are enormous only in America. The actual cost is much lower. American technogy in general is ridiculously overpriced -- compare the cost for raw compute on AWS versus Hetzner for example.

[procgen]

Apples to oranges. The AWS premium buys you a much broader platform with autoscaling, dozens of instance families, managed databases, serverless products, IAM, global regions, etc.

[znnajdla]

Shit that 99% of CRUD app SaaS don't need.

[brap]

I bet that, all things considered, training 10 new Chinese frontier models costs _significantly_ less than a single modern fighter jet.

[well_ackshually]

Let's play a dumb game and assume that training a new model like K3 costs a billion dollars. (It doesn't, but let's play the dumb game still).

If you count it as part of the defense budget, it is 0.3% of their budget. As part of their education budget? 0.05%. Science? 0.5%.

It's pocket change. They can blow a dozen Kimis every year while their own industry is improving and it's a pain in the ass in the US's backside. They're laughing their asses off watching companies having their values inflated to trillions of dollars, more than the GDP of dozens of countries while producing nothing in value more than GPT 5.6.