US Grants Limited Access to Anthropic's Mythos 5 AI Model for Select Cybersecurity Firms
The U.S. government has authorized Anthropic to provide its Mythos 5 AI model to a limited group of trusted American cybersecurity companies, after a prior block by the administration.
Anthropic announced that the United States government has permitted the redeployment of its Claude Mythos 5 artificial-intelligence model to a select set of American cybersecurity firms that protect critical infrastructure. The company said the notification arrived on Friday and that it is working to restore access quickly while continuing discussions with officials to broaden availability of Mythos 5 and to make the related Fable 5 model generally accessible again.
The decision follows a June 12 order that had halted distribution of Mythos 5 and Fable 5, citing concerns that powerful AI systems could be misused by foreign military intelligence. The earlier restriction was imposed by the Trump administration, which did not disclose the criteria for vetting organizations that would receive the technology.
According to a source familiar with the new directive, more than 100 companies and institutions, including several Fortune 500 firms, are expected to gain access. OpenAI CEO Sam Altman expressed reservations about government-selected customers on social media, noting that while extensive safety testing is prudent, he dislikes the idea of the government determining which entities receive advanced models.
Anthropic said it remains in contact with the government to expand access and to eventually return the models to broader commercial use.