The U.S. Department of Commerce has lifted export restrictions on Anthropic's most powerful artificial intelligence models, Claude AI Fable 5 and Mythos 5, less than three weeks after imposing the controls. Anthropic will begin restoring access to both models immediately.
In exchange for the removal of restrictions, Anthropic has committed to cooperating with the U.S. government on artificial intelligence safety matters, identifying potential risks, reporting instances of malicious use, and maintaining ongoing dialogue with authorities ahead of future model releases. The Department of Commerce retained the authority to reinstate export controls if Anthropic fails to meet its obligations or circumstances change materially.
The reversal underscores the evolving regulatory landscape surrounding advanced artificial intelligence development, where government agencies are balancing national security concerns with the need to maintain U.S. competitiveness in the AI sector against international competitors.