Anthropic halts Mythos access as US imposes sweeping foreign-use AI controls

Anthropic has disabled access to its advanced AI models, Mythos 5 and Fable 5, for all users after a U.S. government order directing a broad suspension of foreign-national access to frontier AI citing national-security concerns. Anthropic says the directive, issued by the Commerce Department, is being complied with, even as it argues that a narrow jailbreak finding should not halt deployment of widely used models. The move underscores growing government involvement in AI policy and could set a precedent affecting other major AI developers, with ongoing tensions between security concerns and innovation.