NSA reportedly utilizing Anthropic's Mythos model amid Pentagon tensions

Apr 20, 2026

AI Summary

The National Security Agency is reportedly using Anthropic's Mythos Preview model, which was withheld from public release due to concerns over its capabilities for offensive cyberattacks. This development follows a dispute between Anthropic and the Department of Defense regarding access to the model and its implications for national security.

The NSA is said to be using Mythos Preview, a model developed by Anthropic, for cybersecurity tasks, specifically to scan for exploitable vulnerabilities.
Anthropic announced Mythos earlier this month but limited its access to about 40 organizations, citing concerns over its potential for offensive cyberattacks.
The Department of Defense previously labeled Anthropic a “supply chain risk” after the company restricted access to its model's full capabilities.
The UK’s AI Security Institute has also confirmed access to Mythos.
The Pentagon is currently involved in legal disputes regarding the use of Anthropic's tools, arguing they could pose national security threats.
Anthropic's relationship with the Trump administration appears to be improving, as indicated by a recent meeting between its CEO and White House officials.

anthropicmythosnsaai modelpentagon

NSA reportedly utilizing Anthropic's Mythos model amid Pentagon tensions

Related Stories

Leaders Face Challenges in Managing AI Agents in the Workplace

Concerns Rise Among Lawyers Over Use of AI Note-Taking Tools

Emerging 'Wired Belt' highlights political potential of white-collar workers facing AI job risks

Future Workspaces May Embrace Dictation Technology and Whispering