The United States government has lifted curbs on Anthropic's Fable and Mythos artificial intelligence models, allowing their global release. The move comes after the company agreed to additional safety testing demanded by the Trump administration.
Anthropic had previously spooked the administration into requiring safety evaluations before granting export clearance. The restrictions had delayed the international rollout of these advanced systems, which represent a significant leap in AI capabilities.
The government specifically released the Fable 5 and Mythos 5 models, according to Wired. The approval came with conditions: Anthropic added a new security measure to satisfy regulatory concerns, though the exact nature of that measure remains undisclosed.
This decision signals a shift in the administration's stance toward frontier AI export controls. It may set a precedent for how other firms handle government oversight, potentially accelerating global deployment of cutting-edge models.
Critics argue the government is now relying on a company's internal safety promises rather than independent audits. The long-term effectiveness of such voluntary measures remains uncertain.