The U.S. government has authorized Anthropic to restore access to its Claude Mythos 5 cybersecurity-oriented model for a specified set of more than 100 U.S. government agencies and companies, according to reports from Semafor and Reuters. The permission explicitly includes non‑American employees working at those organizations.
Background
Two weeks after a government ban had prompted Anthropic to pull both its Mythos 5 and Fable 5 models from the market, the administration softened its position by allowing a defined group of trusted organizations to use Mythos 5 again. The original restrictions had stopped non‑U.S. persons from accessing the models; the new directive reinstates access for non‑American staff at the approved organizations, and notably includes Anthropic’s own non‑American employees.
In a letter seen by Semafor, Commerce Secretary Howard Lutnick told Anthropic’s chief compute officer Tom Brown that he had “determined that appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model.”
Fable 5 remains unresolved
The new directive does not appear to address Fable 5. That model is a variant of Mythos 5 that had been more widely released shortly before the ban because it was presented as having stronger protections. Both Mythos 5 and Fable 5 were pulled after security researchers reportedly found ways to bypass the guardrails.
Anthropic’s statement
Anthropic acknowledged progress in a post on X: “Since June 12, we’ve been working closely with the US government to restore access to Claude Mythos 5 and Fable 5. 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, and we’re continuing to work with the government to expand access to Mythos 5 and make Fable 5 available for general use again.”
The company has not provided detailed public information about the exact list of approved organizations, the specific safeguards required, or the timetable for broader re‑release. Anthropic did not immediately respond to media requests for further comment.
Why this matters
The move represents a notable effort by the administration to balance national security and safety concerns about powerful AI models with the operational needs of organizations that protect critical infrastructure. The absence of a clear decision on Fable 5, and limited public detail on safeguards and beneficiaries, leaves significant questions about how broadly such models will be permitted to operate going forward.
