At 5:21pm ET on Friday, June 12, Anthropic received an export-control directive from Commerce Secretary Howard Lutnick ordering the company to cut off Claude Fable 5 and Mythos 5 access for all foreign nationals, a scope that in practice meant disabling both models globally. By that evening, the most capable systems Anthropic has ever shipped were dark.
Fable 5 had been live for three days. Anthropic released it on June 9 after what the company describes as “thousands of hours” of pre-launch red-teaming with the U.S. government, the UK AI Safety Institute, and outside groups, paired with a 30-day customer-data retention policy specific to Fable. The company says the launch capabilities “exceed those of any model we’ve ever made generally available.” Mythos 5, the tier above the prior Opus-class systems, had been quietly in use with government agencies and select corporate partners to harden their own systems, per Al Jazeera.
The stated rationale, reported by The Wall Street Journal and routed through Lutnick’s letter to chief executive Dario Amodei, is a jailbreak vulnerability. Anthropic’s response is unusually pointed for a frontier lab. The company says the technique surfaces only “a small number of previously known, minor vulnerabilities,” and that Fortune has independently reported OpenAI’s GPT-5.5 can find the same vulnerabilities without any bypass at all. “We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people,” Anthropic wrote, adding that applying the standard industry-wide “would essentially halt all new model deployments for all frontier model providers.”
The company chose the word “abruptly disable” to describe its own compliance.
None of this is happening in a vacuum. In February, President Trump ordered federal agencies to stop using Anthropic’s models. In early March, the Pentagon designated the company a “supply chain risk” after Anthropic refused contract language permitting use “for any lawful purpose,” a designation it’s now challenging in federal court. David Sacks, the former AI and crypto czar, has publicly accused the company of “regulatory capture” and called it “woke.” The Fable and Mythos shutdown is the fourth move in a sequence, not the first.
What’s notable is who’s flinching on the administration’s side. Dean Ball, a former Trump AI policy adviser, posted on X that the action looked like “lawfare against Anthropic in particular or extreme national-security hawkery” and called it “simply cartoonish.” Speaking to reporters in Ireland on Sunday, Canadian Prime Minister Mark Carney drew the structural lesson: “The situation we’re in collectively right now with Mythos and Fable is something that can happen with over-reliance on certain models.”
Export controls were built for chips, fabs, and dual-use hardware shipments. They’re now being aimed, with hours of notice, at a software product running inside a U.S. company for users in allied democracies. That instrument has been picked up before, against Toshiba in 1987 and Huawei in 2019, but never against a domestic firm’s own commercial deployment to its own customers. The precedent is what’s new.