Anthropic Suspends AI Models Over US Security Concerns

Anthropic has suspended access to two of its new AI models, Fable 5 and Mythos 5, following national security concerns raised by U.S. authorities. The company was ordered to restrict foreign nationals from using Claude Fable 5, which had just been released. This order also impacted Mythos 5, leading Anthropic to disable both systems for all customers. Officials did not identify a specific flaw but were concerned about a possible way to bypass the models' safety controls. This method, known as "jailbreaking," pushes AI systems beyond their intended limits. Anthropic reviewed the demonstration cited by authorities and found it detected minor software weaknesses. The company argues other publicly available models could find similar flaws. This suspension highlights ongoing questions about government oversight of advanced AI systems.