The Trump administration has partially reversed its recent ban on Anthropic’s advanced AI models, allowing the cybersecurity-focused Mythos 5 to be accessed by over 100 designated U.S. government agencies and companies. This revised directive also permits non-American employees within these organizations, including Anthropic’s own international staff, to utilize the model.

Commerce Secretary Howard Lutnick communicated the decision, stating that “appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model.” The move comes two weeks after the initial ban, which saw both Mythos 5 and Fable 5 models withdrawn from the market due to concerns that their security guardrails could be bypassed. While Mythos 5 is now being made available to entities involved in operating and defending critical infrastructure, the directive does not appear to address the release of Fable 5.

Anthropic confirmed the development, expressing gratitude for the government’s cooperation. The company stated on X, “Today, the government notified us that Mythos 5, our strongest cybersecurity model, can be redeployed to a set of US organizations that operate and defend critical infrastructure. We’re restoring access for these organizations quickly, and we’re continuing to work with the government to expand access to Mythos 5 and make Fable 5 available for general use again.” The company is actively working to restore access for these key organizations and is in ongoing discussions to broaden the availability of Mythos 5 and reintroduce Fable 5 for general use.

Key Takeaways

• The Trump administration has eased restrictions on Anthropic's Mythos 5 AI model, allowing access for over 100 U.S. government agencies and companies.
• Non-American employees within these approved organizations can now access the Mythos 5 model.
• The decision follows an earlier ban prompted by concerns over the model's security guardrails, with ongoing efforts to reintroduce the Fable 5 model.

Editor’s Analysis & Impact

This policy adjustment signals a pragmatic approach by the administration, balancing national security concerns with the operational needs of critical infrastructure. By allowing trusted partners access to Anthropic’s advanced cybersecurity AI, the government acknowledges its utility in defending vital systems. The continued dialogue suggests a recognition of the importance of these AI tools, even as safety and security remain paramount. The eventual reintroduction of Fable 5 could further underscore the evolving regulatory landscape for powerful AI, aiming to foster innovation while mitigating risks.

Frequently Asked Questions

Q: Why was Anthropic's AI model initially banned?
A: The models, Mythos 5 and Fable 5, were initially pulled from the market because security researchers reportedly found ways to easily bypass their guardrails, raising concerns about potential misuse and security vulnerabilities.

Q: Which organizations can now access Mythos 5?
A: Over 100 specific U.S. government agencies and companies, particularly those involved in operating and defending critical infrastructure, are now permitted to access the Mythos 5 model.

Q: What is the status of the Fable 5 model?
A: The recent directive from the Trump administration does not appear to address the release of Fable 5. Anthropic is continuing to work with the government to make Fable 5 available for general use again.