The U.S. Department of Commerce has officially lifted export restrictions on two flagship artificial intelligence models from Anthropic — Claude Fable 5 and Mythos 5. The decision took effect on June 30, and starting July 1, the company is restoring full global access to these systems.
This event marks the end of an almost three-week standoff between Anthropic and U.S. regulators. The restrictions were imposed on June 12, just three days after the models' release on June 9. Authorities cited national security risks related to the potential bypassing of AI safeguards. In response, Anthropic was forced to completely block access to Fable 5 and Mythos 5 for all international users across all its platforms — from Claude.ai and the API to partner services like AWS Bedrock.
What has changed?
Essentially, the ban created a unique precedent: it became impossible to filter users by nationality in real time without completely stopping the service. The company took extreme measures, disabling both models worldwide until the situation was resolved. Now, after intensive negotiations with the White House, the restrictions have been lifted. Commerce Secretary Howard Lutnick confirmed that over two weeks, the parties conducted a detailed analysis of Fable 5 to ensure its compliance with U.S. national security interests.
Why is this important for the market?
Claude Fable 5 is Anthropic's most powerful and widely available model, built on the Mythos architecture. It is optimized for complex tasks, handling large volumes of data, programming, and image analysis. Mythos 5, in turn, is designed for critical and sensitive fields such as cybersecurity and was initially available only to a limited circle of partners through the Project Glasswing program.
Pricing remains competitive: $10 per million input tokens and $50 per million output tokens. Anthropic's internal classifiers automatically redirect high-risk queries (e.g., those related to bio- or cybersecurity) to safer algorithms, which became a key argument in negotiations with regulators.
Cryptalist Analytics: The swift lifting of restrictions is not just a victory for Anthropic but a signal of the formation of a more mature AI regulation system in the U.S. Instead of harsh bans, we are seeing a shift toward dialogue and targeted checks. For the market, this means reduced regulatory uncertainty, which is especially important for projects building infrastructure based on advanced language models. However, the incident highlighted the vulnerability of global AI services to national political decisions — a factor investors should consider when assessing risks.