NSA Adopts Anthropic's Security-Focused AI Model Amid Ongoing Legal Dispute

The National Security Agency is currently testing Anthropic’s new Mythos Preview model, according to a report from Axios. This development occurs despite a public standoff between the AI developer and the Pentagon, and a broader federal directive to halt the use of Anthropic’s services. The company introduced Mythos in early April, highlighting its proficiency in computer security applications.

The agency’s access follows a meeting last week between Anthropic CEO Dario Amodei and White House officials, including chief of staff Susie Wiles. The White House characterized the discussions as productive, though President Trump later told reporters he was unaware of the engagement.

Anthropic provided Mythos Preview to approximately forty organizations for evaluation. Sources indicate the NSA’s use has expanded beyond initial testing within the department. This activity unfolds against a backdrop of ongoing litigation. In March, Anthropic filed lawsuits challenging the Defense Department’s designation of the company as a supply chain risk—a label that triggered the federal usage ban. While one court issued a preliminary injunction against the designation, a separate court declined to remove it, leaving the legal conflict unresolved.

The situation presents a complex picture: a key intelligence agency is evaluating a tool specifically built for security tasks from a vendor the administration has officially sidelined.

Source: Engadget