"They screwed us": Personality clashes sent Anthropic's models offline
TL;DR
Axios reports that Anthropic took its top models Mythos and Fable offline after the Trump administration imposed strict export controls. Officials say the trigger was jailbreak risk and Anthropic's alleged failure to respect a cyber executive order. On June 11, Amazon CEO Andy Jassy warned Treasury Secretary Scott Bessent that the models could be jailbroken. Anthropic counters that it had explicit government approval to deploy Fable.
Nauti's Take
This is not a normal safety spat. It is a warning shot for anyone building AI products in regulated markets: technical approval is not enough when the political relationship catches fire.
If you want government customers, incident communication is part of the product.
Briefingshow
The episode shows that frontier AI access now depends on more than benchmarks: trust, export rules and executive-level communication can decide whether models stay online. Real cyber risk deserves serious technical review. But if access is pulled because of bruised relationships and political optics, the precedent becomes unstable for every major AI lab.
Sources