In brief
- The U.S. government ordered Anthropic to suspend access to its two most advanced AI models, Fable 5 and Mythos 5, citing a national security concern over a potential jailbreak vulnerability.
- Anthropic complied but disputed the finding, arguing the vulnerability is simple and already replicable using other publicly available models like GPT-5.5.
- The company warned the directive sets a dangerous precedent that, if applied industry-wide, could halt all new frontier AI model deployments.
The U.S. government issued an emergency export control directive on Friday ordering Anthropic to immediately suspend access to its two most powerful artificial intelligence models—Claude Fable 5 and Claude Mythos 5, both released just days earlier—for all foreign nationals, including the company’s own employees, citing national security concerns.
The directive bars any foreign national from accessing the models, whether inside or outside the United States. The sweeping scope of the order forced Anthropic to disable the models for its entire customer base to ensure compliance.
The letter did not provide specific details of its national security concern, but the government believes it has become aware of a method of bypassing, or “jailbreaking,” the publicly available Fable 5 model. Mythos 5, which has fewer guardrails and is particularly powerful at discovering cybersecurity exploits, was only available to select partners.
Anthropic disputed the severity of the finding. The company said it reviewed a demonstration of the technique and concluded that the vulnerabilities identified appear relatively simple, and that other publicly available models are able to discover them as well without requiring any bypass.
The company said the government has so far provided only verbal evidence of a potential narrow, non-universal jailbreak—essentially consisting of asking the model to read a specific codebase and fix any software flaws. Anthropic added that it validated that the level of capability on display is already widely available from competing models, including OpenAI’s GPT-5.5.
While complying with the directive, Anthropic said it believes the action sets a dangerous precedent. “If this standard was applied across the industry,” the company wrote, “we believe it would essentially halt all new model deployments for all frontier model providers.”
Access to all other Anthropic models will not be affected. The company said it is working to restore access as soon as possible.
On Saturday, David Sacks—the co-chair of the President’s Council of Advisers on Science and Technology—wrote on X that “a highly credible, trusted partner of both Anthropic and the U.S. government who was testing Fable came forward with a jailbreak of those guardrails. The admin asked [Anthropic CEO Dario Amodei] to fix the jailbreak or de-deploy the model. Dario refused.”
Sacks further claimed that Anthropic’s reaction to the government’s request ran counter to the company’s own public claims regarding the need for AI safety and regulation—a perspective that Amodei again shared this week in a blog post.
“Anthropic prioritized the continued offering of the consumer model over safety,” Sacks wrote. “In reaction, the admin issued the export control. The admin did this reluctantly. It’s been very surprised that Anthropic hasn’t wanted to cooperate with a reasonable safety request (i.e. fixing the jailbreak issue). Anthropic’s reaction is very much at odds with their branding and ethos as a safe AI research community.”
Sacks added that the administration hopes that Anthropic will fix the issue and that Fable 5 can return to public release.
“The admin values Anthropic’s technical capabilities and feels that this issue, while serious, should be easily resolved,” he wrote. “The ball is in Anthropic’s court.”
Earlier this year, Anthropic and the U.S. government butted heads when the AI firm refused to sign an expanded agreement allowing for mass domestic surveillance of Americans and fully autonomous lethal weapons systems.
President Donald Trump slammed the company after talks broke down, and the Department of Defense labeled Anthropic a “supply chain risk”—a designation that the firm has challenged in court. Since then, there have been reports that the standoff between the two sides has cooled as the government sought use of Claude Mythos and other models.
Daily Debrief Newsletter
Start every day with the top news stories right now, plus original features, a podcast, videos and more.
