After spooking Trump into safety testing, Anthropic AI models get global release (arstechnica.com)

The U.S. government has lifted export restrictions on Anthropic's Claude models, Fable 5 and Mythos 5, following a review initiated by concerns raised during the Trump administration about potential national security risks. Anthropic confirmed that Fable 5 is now globally available, while access to Mythos 5 was restored for U.S. organizations on June 26. The company has committed to expanding access to Mythos through the Glasswing program, which allows trusted cybersecurity researchers to utilize the model for defensive purposes. This development is significant for the AI and ML community as it marks a collaborative effort between Anthropic and the U.S. government to address safety concerns surrounding advanced AI models. Anthropic has been proactive, establishing a program for "red-teaming" its models to identify vulnerabilities and setting up an internal team to continuously monitor emerging threats. Despite the optimistic tone of the announcement, government officials have indicated that the ability to reinstate export restrictions remains a possibility, highlighting the ongoing scrutiny surrounding AI technologies and their implications for national security.