Anthropic's AI Exports Resumed Amid Security Concerns

The U.S. Commerce Department has lifted export controls on Anthropic's Fable and Mythos AI models, reversing a suspension due to national security risks. The decision, allowing releases to trusted organizations, follows increased government oversight of advanced AI models to prevent potential misuse by foreign intelligence agencies.

The U.S. Commerce Department has reversed its decision to suspend exports of Anthropic's advanced AI models, Fable and Mythos, following heightened concerns regarding national security risks. Less than three weeks after the initial order, Anthropic is now permitted to release its Mythos 5 model to a select group of trusted U.S. organizations.

This comes amid the U.S.'s increasing scrutiny of new AI model releases, given their potential misuse by foreign military intelligence agencies, particularly those from China and Russia. Anthropic had previously halted access to its AI models in compliance with the export-control order issued on June 12.

While the reversal of the export ban has been welcomed, it has also sparked criticism about the government's vetting process for companies permitted to access these AI models. OpenAI's CEO Sam Altman expressed concerns over the government's role in determining clients and cited similar issues surrounding the delayed public launch of GPT-5.6 at the behest of government directives.