Despite blacklist, NSA is reportedly using Anthropic’s Mythos: Report

The National Security Agency (NSA) has reportedly begun utilizing Anthropic's latest cybersecurity AI model, Claude Mythos, despite existing government restrictions and the Department of Defense's classification of it as a 'supply-chain risk'. According to a report by Axios, the NSA is one of the approximately 40 organizations granted access to Mythos, although Anthropic has not publicly disclosed the identities of these users. If these allegations hold true, it appears that the government is placing a higher priority on its cybersecurity needs than on its ongoing disputes with Anthropic. Previously, the Pentagon had requested that government personnel refrain from using Anthropic's technology, highlighting a contentious relationship aggravated by legal challenges surrounding the ban on the company’s AI systems. Amidst this backdrop, the military is expanding its use of Anthropic’s tools, including Claude Mythos. However, specifics regarding the NSA's application of the model remain unclear. According to available information, Mythos could potentially be employed for scanning systems to identify security vulnerabilities and weaknesses. Access to Mythos is currently restricted due to the risks associated with its capabilities, with only about 40 organizations being able to use it. Anthropic had initially indicated that 12 organizations would have access, and it is now believed that the NSA is among these undisclosed entities. Notably, the UK also has access through the AI Security Institute. On April 17, Anthropic CEO Dario Amodei met with White House Chief Susie Wiles and Treasury Secretary Scott Bessent to discuss the government's adoption of Mythos, its security strategies, and potential future collaborations. This meeting might pave the way for other government agencies, beyond the Pentagon, to leverage the AI model. While Mythos is designed to enhance cybersecurity efforts, there are ongoing concerns about its potential misuse if it were to fall into the wrong hands. Due to its advanced capabilities, Anthropic is proceeding cautiously, continuing to evaluate safety protocols before considering a wider public release.