The NSA is using Anthropic's new Mythos Preview AI model despite the ongoing feud between the company and the Pentagon. According to Axios sources, NSA is among approximately 40 organizations given access to Mythos Preview, which Anthropic describes as "strikingly capable at computer security tasks." This development comes just months after Trump ordered all government agencies to stop using Anthropic's services.

The news follows a meeting between Anthropic CEO Dario Amodei and White House chief of staff Susie Wiles, reportedly to discuss Mythos. While the White House called the meeting "productive and constructive," President Trump claimed he had "no idea" about it. This apparent contradiction highlights the complex relationship between the AI company and the Trump administration despite ongoing legal battles.

Anthropic remains embroiled in legal challenges with the US government after being labeled a "supply chain risk." The company filed lawsuits against the Department of Defense and won a preliminary injunction in one court, though federal judges in another denied its motion. The continued use of Mythos by the NSA suggests the AI's security capabilities may outweigh political tensions in certain government applications.