The US government has ordered Anthropic to restrict foreign access to its newest AI models, Fable 5 and Mythos 5. The company has shut down both models globally.
According to Oxu.Az, the decision is based on an export control directive. The directive prohibits all foreign nationals, including Anthropic's foreign employees, from accessing the models.
Anthropic stated: "The US leadership issued a directive based on national security authority. As a result, we had to urgently disable both models for all users."
The company received the directive on the evening of June 12. The letter contained no specific details about the threat. The government only verbally informed about a bypass method for the models' protective restrictions.
Anthropic emphasized that this limited vulnerability is not universal and does not justify recalling a commercial model used by hundreds of millions.
The organization warned: "If this approach is applied industry-wide, the release of new models will effectively cease."
Founded in 2021 by former OpenAI employees, Anthropic has become a key player in AI safety. The company launched its Claude chatbot in March 2023.
In March 2026, the US Department of Defense declared Anthropic a "supply chain threat" and cut business ties, following a dispute over Pentagon's intention to use its AI products for all lawful purposes.
