Any AI safety policy that believes in the scaling hypothesis and wants to stop China from poaching the crown jewels of US AI research has likely already failed.
