Anthropic suspended its latest AI models after Commerce Secretary Howard Lutnick directed the company to restrict access for foreign nationals, citing national security risks. The directive came after government officials discovered a jailbreak that could bypass built-in safety measures. The company withdrew Mythos 5 and Fable 5 from all users, including domestic customers, effectively shutting down access entirely.

The Trump administration had been testing the Mythos model for weeks and raised concerns about potential exploitation of cybersecurity vulnerabilities in critical infrastructure like the global financial system. While federal agencies experimented with the technology ahead of release, the administration initially allowed commercial deployment. President Trump recently signed an executive order creating a voluntary inspection framework rather than granting outright blocking authority.

Anthropic faces mounting pressure from Washington, including a Pentagon designation as a "supply-chain risk" amid disputes over military applications. The company previously sued the government over this classification. Former AI coordinator David Sacks and other administration officials have criticized Anthropic's safety policies. Trump has publicly derided the company as run by "leftwing nut jobs."

The suspension highlights growing regulatory tension in the AI sector. Anthropic argues the government's action lacks transparency and technical grounding, calling it a "misunderstanding" that could halt all frontier model deployments industry-wide. The legal battle continues as AI regulation intensifies.