The Trump administration plans to lift export restrictions on Anthropic's Fable 5 AI model as soon as Tuesday evening, according to a U.S. official and multiple sources. The move reverses a 18-day-old suspension that had barred foreign nationals from accessing the powerful Mythos-class system.
Last week, the White House permitted Anthropic to restore the model for a select group of government-approved organizations. The full lift follows weeks of discussions between the company and Commerce Department officials over national security risks.
Commerce Secretary Howard Lutnick stated on X that his office "worked closely with Anthropic to analyze and approve Fable 5 to ensure alignment across the US Government." It remains unclear what technical or policy adjustments the firm made to address concerns.
The decision restores public access to a frontier AI system but leaves broader questions unanswered. The U.S. government's role in regulating advanced models before release remains ad hoc, creating uncertainty for AI developers navigating export rules.
Critics argue the flip-flop shows the administration lacks a coherent framework for evaluating AI risks. Without clear standards, future security reviews risk becoming reactive rather than systematic.