The Trump administration has lifted export controls on Anthropic’s Fable 5 and Mythos 5 AI models following a deal with the Commerce Department. The decision was announced in a letter from Commerce Secretary Howard Lutnick to Anthropic cofounder Tom Brown, viewed by WIRED. The move removes the need for licenses to export, reexport, or transfer these models domestically, including deemed exports. This marks a significant shift in the regulatory landscape for Anthropic’s most advanced AI systems.
Anthropic has been collaborating with the Commerce Department and the White House to enhance safeguards against users bypassing Fable’s safety restrictions to access restricted capabilities, particularly in cybersecurity. According to sources, the company has agreed to proactively detect and address security risks tied to the models. Lutnick also highlighted that Anthropic will work closely with the U.S. government on protocols and standards for Mythos, Fable, and future models. These measures aim to align with the administration’s security priorities while enabling broader model access.
The decision follows a months-long negotiation between Anthropic and the Trump administration, led by Commerce Secretary Lutnick and national cyber director Sean Cairncross. Initially, Anthropic argued that the administration’s security concerns were overstated, claiming it was impossible to guarantee zero jailbreaks for the Mythos model. However, the company shifted its approach, focusing on implementing stronger safeguards to reduce jailbreaks rather than debating the feasibility of preventing them entirely. This strategic pivot reflects a broader effort to satisfy regulatory requirements and resume model operations.
Source: wired
