Axios he says that the NSA is using Anthropic's limited Mythos Preview model, despite the Pentagon insisting that the company poses a "supply chain risk." 
The government's cybersecurity needs appear to outweigh the Pentagon's dispute with Anthropic. The military is now expanding its use of Anthropic's tools, while at the same time arguing that the use of those tools threatens U.S. national security.
Two sources said the NSA uses Mythos, while another said the model was also used more broadly within the department. It is not known how the NSA currently uses Mythos, but other organizations with access to the model use it primarily to scan their own environments for exploitable security vulnerabilities.
Anthropic limited access to Mythos to about 40 organizations, arguing that its offensive cyber-dynamic capabilities were too dangerous to allow wider release.
Anthropic announced only 12 of these organizations.
One source said the NSA was among the organizations with access that Anthropic did not report.
The NSA's counterparts in the UK have said they have access to the model through the country's Artificial Intelligence Security Institute.
Anthropic's CEO met with top U.S. officials on Friday to discuss "opportunities for collaboration," according to a White House spokesman, "as well as common approaches and protocols to address the challenges associated with scaling this technology."
Although the press releases will range from very select to rare, I said I'd pass...because sometimes the editors hide.
“Following an order from the Trump administration on February 27, 2026, the Pentagon is phasing out the use of Anthropic AI products, with a 180-day deadline (approximately late August 2026) for their complete removal.”