According to a report by Axios, the National Security Agency (NSA) is reportedly utilizing Mythos Preview, a sophisticated model developed by Anthropic that the company deliberately kept from public distribution. This development surfaces mere weeks after the Department of Defense (DoD), the NSA's overarching agency, designated Anthropic as a "supply chain risk." This classification stemmed from Anthropic's refusal to grant Pentagon officials unfettered access to the complete functionalities of its AI model.
Anthropic introduced Mythos earlier in the month, characterizing it as a cutting-edge model specifically engineered for cybersecurity applications. However, the company asserted that the model possessed such significant capabilities for offensive cyber operations that it could not be released to the general public. Consequently, the AI company restricted access to Mythos to approximately 40 entities, publicly disclosing the identities of only twelve of them.
The NSA reportedly numbers among these undisclosed recipients, with its primary application of Mythos believed to be the systematic scanning of digital environments for potential vulnerabilities that could be exploited. Furthermore, the UK’s AI Security Institute has independently verified its own access to the Mythos model.
The increasing adoption of Anthropic’s tools by the U.S. military unfolds concurrently with the company's ongoing legal arguments that these very tools could pose a threat to national security. The genesis of the Pentagon’s disagreement with Anthropic traces back to the company's refusal to permit the use of its Claude model for widespread domestic surveillance and the development of autonomous weapon systems.
The NSA’s procurement of Mythos coincides with what appears to be a notable improvement in Anthropic’s relationship with the Trump administration. Just last Friday, Dario Amodei, Anthropic’s chief executive, held discussions with Susie Wiles, the White House chief of staff, and Scott Bessent, the Secretary of the Treasury.
Reports indicate that the White House characterized this recent meeting as productive.
TechCrunch made attempts to solicit comments from the NSA regarding these developments. Anthropic, however, chose to refrain from providing a statement.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.