The US government has issued an export control directive to suspend access to Anthropic Fable 5 and Mythos 5, citing national security concerns. Consequently, the company must disable these artificial intelligence models for all customers immediately. The directive, received on June 12, 2026, at 5:21 PM Eastern Time, applies to all foreign nationals, including foreign employees of the firm.
While access to other models remains unaffected, this sudden suspension disrupts services globally. Meanwhile, the official letter did not provide specific details regarding the exact national security threat. However, the developer understands the decision relates to a potential method of bypassing, or jailbreaking, the model’s safety protocols. As a result, the company had to act quickly to ensure full legal compliance.
Disabling Anthropic Fable 5 for Compliance
In an official statement, the company expressed disagreement with the regulatory action.
“We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people.”
Anthropic Statement
Technical Details of the Alleged Vulnerability
Notably, the organization had previously worked with the US government and the UK Artificial Intelligence Safety Institute to test the safeguards of Anthropic Fable 5. These extensive red-teaming efforts spanned thousands of hours prior to the official launch. Furthermore, the firm implemented a defense in depth strategy, combining narrow safeguards with a 30-day data retention policy to monitor potential cybersecurity risks.
Government Evidence and Model Capabilities
Specifically, the government provided verbal evidence of a narrow jailbreak involving code analysis. The technique reportedly asks the model to read a codebase and fix software flaws. However, the company validated that this capability is widely available in other public models, including OpenAI’s GPT-5.5, which software defenders use daily. Consequently, they argue the vulnerability does not present a unique risk.
Industry Implications and Next Steps
The developer warned that applying this regulatory standard broadly would halt new deployments of Anthropic Fable 5 and other models. Nevertheless, the company is complying with the legal directive while working to resolve the misunderstanding. In addition, the team plans to share more technical details within the next 24 hours to clarify the situation.
