Washington Can Suspend Anthropic’s Models, but It Can’t Fix Your Architecture

In June, the US Department of Commerce ordered Anthropic to suspend access to its Fable 5 and Mythos 5 AI models for foreign nationals due to national security concerns linked to a jailbreak exploit. Anthropic temporarily cut off access entirely, then restored it weeks later, highlighting challenges in AI governance amid geopolitical pressures. The incident underscores ongoing risks of AI models being used for offensive cyber activities and the complexities organizations face in managing AI risk