Governments can now switch off frontier AI models, making AI infrastructure a sovereign risk...
Governments can now switch off frontier AI models, making AI infrastructure a sovereign risk that boards cannot ignore. The US government used export control authority to order Anthropic to take its frontier AI models (Fable and Mythos) offline.
Why it matters
This is the strongest daily candidate from 2 distinct sources in the current research run.
Sources
- AI stack is now critical infrastructure — and a government just proved it by switching off a frontier modelThe AI Optimist - 2026-06-19
- A government just switched off a frontier AI model — your AI stack is now critical infrastructureThe AI Optimist - 2026-06-19
- Trump Administration Forces Anthropic's Newest AI Models Offline via Export Control OrderTechCrunch AI - 2026-06-21
- U.S. export controls on Anthropic's AI models Fable and Mythos: a test case with historical parallelsTechCrunch AI - 2026-06-19