The Chonkerton

"They screwed us": Personality clashes sent Anthropic's models offline

ai

According to tech journalist Simon Willison, Anthropic's Fable and Mythos models remain offline following a US government export control directive. The impasse centers on AI safety: the government wants jailbreak-proof models, while Anthropic says perfect jailbreak resistance may be impossible. Personality clashes between company leadership and the administration have added friction to negotiations. High-level meetings at the Commerce Department are happening this week, but the proposed resolution—that everyone involved simply feel more 'safe, secure, and happy'—suggests the two sides are quite far apart.

Source: https://simonwillison.net/2026/Jun/15/axios-clashes-anthr...

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton