US Government Orders Anthropic to Halt AI Models Over Security Concerns

On June 13, 2026, the U.S. government issued an emergency export control order requiring Anthropic to immediately suspend access to its two most advanced artificial intelligence models, Claude Fable 5 and Claude Mythos 5. The directive, prompted by national security concerns, restricts access to these models for all foreign nationals, including Anthropic's own employees, regardless of their location.

This sweeping measure forced Anthropic to disable the models for its entire customer base to comply with the order, which lacked detailed explanations on specific security risks. The government reportedly expressed concerns about a potential risk of a so-called "jailbreak" that could allow unauthorized access to the functionality of the Fable 5 model, which was made publicly available just days before the order.

Mythos 5, known for its capabilities in identifying cybersecurity exploits, had limited availability to select partners. In response to the government's directive, Anthropic criticized the assessment as an overreaction, arguing that the vulnerabilities highlighted by the government are relatively straightforward and can also be found in other publicly available models, such as OpenAI's GPT-5.5.

Anthropic pointed out that the government provided only verbal evidence regarding the existence of a narrow, non-universal jailbreak situation, primarily involving a technique for the model to analyze specific code and rectify flaws. Anthropic emphasized that similar levels of competency are already available among competing models without the need for any form of unauthorized access.

The company expressed increasing concern that enforcing such standards across the industry could set a dangerous precedent. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier AI technologies," Anthropic warned.

This situation raises significant questions about the balance between national security and fostering innovation in the AI field. As regulations around AI continue to evolve, stakeholders will be closely monitoring how such directives influence future developments and deployments of advanced AI technologies.