Anthropic disables two AI models after federal export control order

Anthropic disabled access to its newest AI models, Fable 5 and Mythos 5, on Friday after receiving a federal export control order. The order required the company to block foreign nationals from using the models. The directive came three days after the models were released.

Background on the models Fable 5 was released to the public with safeguards against uses the company considered dangerous. A new version of Mythos was provided to a small group of cyber-defenders and infrastructure providers with fewer guardrails. Both models are based on the Mythos model that the company initially chose not to release to the broader public over concerns it could supercharge hacking capabilities.

Industry reaction Several outlets reported that Amazon had reached out to the administration to flag concerns about a potential method of bypassing the guardrails on Fable. A source close to Anthropic said the company was given 90 minutes to pull Fable without prior indication of a national security threat.

Brad Carson, president of Americans for Responsible Innovation, said the action showed that ad hoc executive actions were replacing clear standards. Anthropic sent senior technical staff to Washington to meet with administration officials after the models were pulled.

The staff have been meeting virtually with officials daily since the administration first contacted the company.