The Trump administration has formally lifted export controls on Anthropic’s advanced AI models Mythos 5 and Fable 5, following an agreement with the company to implement stronger safeguards aimed at reducing cybersecurity risks and model misuse.
- License restrictions lifted on Mythos 5 and Fable 5 AI models
- Anthropic agrees to enhanced safeguards against security risks
- Deal resolves months-long dispute between company and Commerce Department
What happened
The Trump administration’s Commerce Department announced it has lifted export controls on Anthropic’s two most powerful AI models, Mythos 5 and Fable 5. Previously, these models were restricted to select companies and government agencies only, requiring licenses for any export or transfer. The change was communicated in a letter from Commerce Secretary Howard Lutnick to Anthropic cofounder Tom Brown.
This decision follows several weeks of negotiations during which Anthropic pledged to strengthen its internal safeguards to prevent unauthorized access to restricted functionalities of these AI models. The company committed to proactively detecting and addressing any security vulnerabilities associated with the models.
Why it matters
The lifting of export controls marks a significant easing of regulatory constraints on cutting-edge AI technologies, reflecting a balance between innovation and national security concerns. Anthropic’s agreement to collaborate with the government on security standards underscores the growing complexity of managing advanced AI capabilities in a global environment.
The dispute highlighted challenges around controlling AI misuse, such as jailbreaks that could unlock powerful but potentially harmful model features. By resolving these issues, the administration aims to ensure that advanced AI can be shared more broadly without compromising cybersecurity or safety.
What to watch next
Future developments will focus on how effectively Anthropic implements its promised safeguards and cooperates with the U.S. government on ongoing security protocols. Observers will watch whether these measures can minimize risks of jailbreaks or other exploits in practical use cases.
Additionally, the move could set a precedent for how other countries regulate AI exports, influencing international standards and the competitive landscape for advanced AI technologies. Industry participants will be keen to see if other high-profile AI models face similar export controls or if U.S. policy shifts toward wider access.