The Trump administration forced Anthropic to take its most powerful artificial intelligence models offline just days after their release, citing national security concerns linked to a suspected China-connected firm.
The Commerce Department, in a letter signed by Secretary Howard Lutnick, ordered Anthropic to suspend all access to two models, Fable 5 and Mythos 5, for any foreign national anywhere in the world.
The restriction applied broadly, covering foreign nationals both inside and outside the United States, including employees of Anthropic itself who hold foreign citizenship.
Anthropic had notified the government multiple times ahead of the June 9 release of Fable, a general-use version of its more powerful Mythos model, and received no objection from officials before launch.
The situation escalated rapidly on Thursday evening when Amazon contacted senior administration officials to share a report demonstrating how portions of the Mythos model could be jailbroken in ways that posed a national security threat.
At least five other companies also reached out to a range of senior administration officials Thursday evening and Friday morning, adding further pressure on the White House to act against the newly released models.
Administration officials spent hours on Friday speaking directly with Anthropic, urging the company to voluntarily pull its latest model offline, but those efforts failed to produce an agreement.
By Friday night, the model had been taken down, marking a sharp and public reversal for a company that had positioned Fable as its most significant commercial release to date.
The episode revealed the extent to which Anthropic had lost standing in Washington, with the sequence of events suggesting the company no longer controlled the terms of how its most advanced systems could be deployed publicly.
Anthropic has since launched an intensive outreach effort targeting senior administration officials, as the company works to rebuild trust with the White House following the forced restrictions on its flagship models.
The fallout has also cast a shadow over Anthropic’s planned public listing, with the sudden government intervention raising questions among investors and industry observers about the company’s regulatory relationships and market position.
The Commerce Department’s use of national security authorities to demand the suspension represents one of the most direct government interventions in a major AI product launch seen from a leading American artificial intelligence company.
The incident underscores the growing tension between the rapid commercial deployment of frontier AI systems and the national security calculations driving policy in Washington, a conflict likely to shape how advanced AI models are released going forward.
