>If the end goal is that only regulated US companies can use Fable, that is a pretty good outcome for Amazon

It's a terrible outcome for Amazon because it destroys Anthropic's revenue. Roughly half of Anthropic's customers are foreigners, and they wouldn't use Anthropic if its next generation model was banned while other providers' next generation models aren't. And if the US follows through and bans all Mythos-level models for foreigners, then in 6-12 months the entire global market will be overtaken by China when its models catch up, and Amazon will lose money on its investment in OpenAI too.

Immediate revenue impact is basically 0 - nobody cancels their Claude sub because Fable isn’t why they got it in the first place (by nobody I mean like 1% of total users and they’re likely net neutral tokenmaxxers for revenue).

Signal to OpenAI and Google is clear: can’t release too smart models or they get controlled. It follows there is no danger to revenue since other providers are forced to plateau at the same level.

…which puts the whole train the next model business idea a risky proposition since the training can’t ever pay for itself - but USG really wants you to keep training, so guess what happens?

Oh and re China - if you think they’ll release an open Mythos-class model, I have a bridge to sell.