Anthropic Suspends New AI Models Following Government Directive
- •Anthropic has suspended access to its latest models, Claude Fable 5 and Mythos 5, for all customers.
- •The US government ordered the restriction of access for foreign nationals, citing national security concerns.
- •The suspension occurred just 3 days after release, raising questions about AI model export control standards.
On June 12, 2026 (local time), Anthropic announced the immediate suspension of its latest models, Claude Fable 5 and Claude Mythos 5, for all customers. This action follows a US government export control directive that prohibits foreign nationals from accessing both models, based on national security authority. The models had been released on June 9, meaning they were available for only 3 days before the shutdown.
Claude Fable 5 represented the company's highest-performing "Mythos-class" model, demonstrating advanced capabilities in software engineering and scientific research. Mythos 5 was intended for limited distribution to select clients involved in critical infrastructure protection, with certain safety safeguards removed. Anthropic stated that disabling both models was necessary to comply with the directive, noting that no other public models are affected.
According to Anthropic, the government directive was received at 5:21 PM ET on June 12, without providing specific details regarding national security concerns. The company speculates that the government may have had concerns regarding "Jailbreak" attempts against Fable 5. Anthropic contends that it implemented appropriate safety measures, including thousands of hours of Red Teaming.
The suspension occurred shortly after Anthropic CEO Dario Amodei (co-founder and CEO of Anthropic) published an essay arguing for the necessity of government AI policy frameworks. While acknowledging that government intervention in AI deployment is necessary, Anthropic emphasized that such regulations should be clear, fair, and based on technical facts. The focus now shifts to the extent to which the government will clarify the basis for its directive and how AI models will be managed as national security assets, similar to semiconductors.