U.S. Government Approves Limited Release of Anthropic’s Powerful Mythos 5 AI Model
In a significant easing of tensions between the Trump administration and Silicon Valley, the U.S. government has granted AI firm Anthropic permission to release its potent Claude Mythos 5 AI model to a select group of approximately 100 companies and federal agencies.
The announcement, confirmed by CNBC on June 26, 2026, comes in the wake of the Commerce Department’s export block on Mythos 5 and its sibling model, Fable 5. This action was taken due to national security concerns, with authorities fearing that the advanced cybersecurity-focused model could be exploited by malicious entities to identify and exploit software vulnerabilities at an unprecedented pace.
U.S. Commerce Secretary Howard Lutnick wrote in a letter to Anthropic: “I have determined that appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model.” The government emphasized its diligent efforts over two weeks to ensure America’s continued global leadership in AI while safeguarding national security.
The journey has been tumultuous for Anthropic. The company was previously labeled a “supply chain risk” by the Department of Defense — a designation typically reserved for foreign adversaries — after negotiations over military use of its models fell through. In response, Anthropic initiated a lawsuit against the Trump administration, and the litigation is still ongoing.
Interestingly, the government’s letter did not approve the reinstatement of the less powerful Fable 5 model, which was briefly the most advanced AI model widely available to consumers. Talks regarding Fable 5’s reinstatement are reportedly ongoing. This incident is widely viewed as the inception of a new U.S. regulatory framework that grants the federal government direct control over the release of frontier AI models.
Source: CNBC — Trump admin allows Anthropic to release Mythos AI model to select entities