Trump Administration Permits Anthropic's Claude Mythos 5 for 100+ US Organizations After Two-Week Ban

The Trump administration is allowing Anthropic to deploy its Claude Mythos 5 model to more than 100 specific US government agencies and companies, two weeks after banning the cybersecurity-oriented model. Commerce Secretary Howard Lutnick approved the limited release in a directive sent to Anthropic's chief compute officer Tom Brown on Friday.

What Changed

The June 12 ban forced Anthropic to pull both Mythos 5 and Fable 5 from the market after security researchers allegedly bypassed the models' guardrails easily. The new directive specifically addresses Mythos 5 only, leaving Fable 5's status unchanged.

"I have determined that appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model," Lutnick wrote, according to documents seen by Semafor.

The approved organizations include those operating and defending critical infrastructure. Notably, the directive permits non-American employees at these organizations to access the model—a reversal from the original ban that blocked non-Americans, including Anthropic's own international staff.

What Anthropic Says

Anthropic acknowledged the development in a post on X: "Today, the government notified us that Mythos 5, our strongest cybersecurity model, can be redeployed to a set of US organizations that operate and defend critical infrastructure. We're restoring access for these organizations quickly."

The company stated it is "continuing to work with the government to expand access to Mythos 5 and make Fable 5 available for general use again."

Background on the Ban

Mythos 5 is described by Anthropic as its strongest cybersecurity model. Fable 5, a version released days before the ban, was marketed as having additional protections. Both were withdrawn after reports that their safety guardrails could be easily circumvented.

The ban prohibited non-Americans from accessing the models, creating operational challenges for Anthropic and its customers with international workforces.

What This Means

The partial reversal signals the Trump administration is attempting to balance AI safety concerns with practical deployment needs for critical infrastructure protection. The selective approval—limited to specific organizations and excluding general availability—suggests continued caution around powerful cybersecurity models.

The directive's silence on Fable 5 indicates the government may view it as having insufficient safeguards despite Anthropic's claims of enhanced protections. The timeline for broader access to either model remains unclear, leaving most Anthropic customers without access to these cybersecurity capabilities.