More than 70 cybersecurity professionals have signed an open letter urging the U.S. government to reverse an export control order that blocks access to Anthropic's most advanced AI models, Fable and Mythos. The letter, published Monday and signed by 76 experts so far, warns that the restrictions have stripped defenders of the best tools available for finding software vulnerabilities at a time when foreign adversaries are rapidly advancing their own cyber capabilities. "To pull the best capabilities away from defenders without a good reason when our adversaries are rapidly advancing is dangerous," the letter reads, adding that the order "has taken the best models away from defenders."

The order, issued Friday and reported by Anthropic the same day, cited unspecified national security concerns and gave the company no detailed explanation for the action. In response, Anthropic suspended access to Fable and Mythos for all users worldwide, not just those outside the United States. The signers include some of the most recognized names in the field: former Facebook chief of security Alex Stamos, Bugcrowd founder Casey Ellis, cryptographer and former Apple security architecture manager Jon Callas, computer scientist Paul Vixie, former Block head of applied security engineering Dino Dai Zovi, Luta Security founder Katie Moussouris, and SocialProof Security CEO Rachel Tobac.

The controversy highlights the tension between Anthropic's own cautious rollout of the models and the U.S. government's stricter approach. When Mythos debuted as a preview in April, Anthropic restricted access to roughly 50 companies, citing its exceptional ability to identify security flaws and the risk that malicious hackers or foreign governments could weaponize it. The company has since expanded access to about 150 organizations across 15 countries. Last week, Anthropic released Fable, a public-facing version of Mythos with guardrails designed to block prompts related to biology, chemistry, and cybersecurity, as well as attempts to distill the model and recreate it. In practice, those guardrails were so restrictive that cybersecurity researchers found Fable effectively useless for any meaningful security work. Anthropic has suggested the White House order may have been based on a report flagging concerns about