• The US government has allowed Anthropic to release its Mythos 5 AI model to over 100 trusted partners.
• This decision comes after previous restrictions due to national security concerns regarding misuse.
• Companies included in the rollout mainly consist of Fortune 500 firms and cybersecurity organizations.
• Restrictions remain for those not on the approved list, as the rollout marks a partial reversal of earlier measures.

The U.S. government has authorized Anthropic to release its powerful Claude Mythos 5 AI model to over 100 "trusted partners," which include many Fortune 500 companies and cybersecurity firms. This decision follows a two-week suspension related to national security concerns, particularly fears over the misuse of AI technology by foreign military intelligence, according to Reuters and Channel News Asia.

According to a letter from U.S. Commerce Secretary Howard Lutnick, significant progress has been made in addressing the risks associated with Mythos 5. "I have determined that appropriate safeguards are in place," Lutnick stated, while noting that the export license requirement has been lifted for trusted companies and their non-U.S. citizen employees. However, restrictions still apply to companies not on the approved list, as highlighted by India Times and South China Morning Post.

The decision marks a reversal of restrictions imposed earlier in June, when Anthropic was required to disable access to both Mythos 5 and another model called Fable 5 due to concerns about safeguarding against exploitation by malicious actors. Following discussions with the U.S. government, Anthropic confirmed that it is "restoring access quickly," indicating a commitment to expanding access further and potentially making Fable 5 available to a broader public in the future, as reported by Reuters and India Times.

However, the implementation of these permissions has drawn criticism and raised questions regarding transparency in the selection process for trusted partners. Critics highlight concerns that the government is exercising too much power in vetting which companies gain access to crucial AI models, potentially fostering a lack of transparency in decisions that affect other organizations, according to South China Morning Post.