Anthropic's two most powerful AI models had barely been on the market for three days when the U.S. government told the company to shut them down for foreign users. Fable 5 and Mythos 5 launched June 9th; by June 12th, according to The Verge, a federal order required Anthropic to block international access, citing national security concerns.
The move didn't emerge from nowhere. According to Livemint, Defense Secretary Pete Hegseth had sought to designate Anthropic a supply chain risk — described as an unprecedented action against a domestic U.S. company. TechCrunch reports the administration's decision forced Anthropic to pull its latest cybersecurity models, and suggests the order could be "reactionary, retaliatory, or both," framing it as a signal that "the AI industry isn't immune from U.S. government interference."
Fortune traces the spark to a deceptively mundane incident: three words — "Fix this code" — that allegedly triggered government alarm over what the models could do in the wrong hands. The Pentagon's prior dispute with Anthropic added further friction, and Breaking Defense examined how the Commerce Department crackdown could ripple through military AI programs.
The fallout spread quickly. Microsoft, according to Memeburn, blocked its own employees from using Claude Fable 5 over data retention concerns. Cybersecurity executives, as reported by U.S. News & World Report, publicly urged the Trump administration to ease the restrictions, warning of operational harm. Anthropic sent employees to Washington to meet with White House officials, according to UPI, as Forbes described tensions around Claude Mythos reaching "fever pitch."
International observers took note too. According to the New Indian Express, India's commentators cited the episode as a warning that AI is now a geopolitical tool — one nations must help shape, not merely consume. Meanwhile, the Times of India reported that China's Defense Ministry issued its own warning that excessive AI use in the military risks spiraling out of control.
The episode matters because it shows that even a well-funded, safety-focused American AI company can have its flagship products switched off by executive action — raising urgent questions about how policy risk now sits at the center of every enterprise AI strategy.