This isn’t a standard procurement dispute. According to Tech Policy Press, Secretary of Defense Pete Hegseth designated Anthropic a “supply chain risk”, a designation that, under the DoD’s interpretation, prohibits companies with defense contracts from working with Anthropic altogether, not only from using its technology on federal government systems. The reach is broader than it first appears.
Anthropic is seeking to overturn the designation in court. The core of the dispute is a set of assurances Anthropic sought from the Department: that its technology not be used for mass domestic surveillance of US citizens, and that it not be deployed in autonomous weapons. The DoD refused. Chicago Council on Global Affairs analyst Suzanne Nossel, writing on March 17, framed the standoff as exposing “two faces” of Claude, the version with Anthropic’s constitutional ethical constraints, and a version the Pentagon wants available without them.
A Pentagon memo reportedly issued in early March 2026 is said to have formalized the designation, with a reported 180-day deadline for removal. Neither the specific memo date nor the 180-day timeline has been confirmed via accessible primary sources, treat these as reported, not confirmed.
What is confirmed: the designation is real, the court challenge is real, and the contractor scope is broader than a typical federal vendor dispute. Tech Policy Press reports that European capitals are watching closely, reading the designation as a signal for their own AI sovereignty and defense procurement frameworks.
The practical question for compliance teams and AI vendors isn’t whether this applies to them today. It’s whether the mechanism being established, supply chain risk designation of an AI lab, creates a template other agencies or governments could use.