AI Wiki · Mediocre Security
Legal action

Anthropic v. Department of Defense

Anthropic's challenge to its Pentagon "supply chain risk" designation, rejected 2-1 by the D.C. Circuit on Sep 25 2026.

In February 2026 the Pentagon designated Anthropic a “supply chain risk” after Anthropic refused to drop usage limits on mass surveillance and autonomous weapons. On Sep 25 2026 the D.C. Circuit ruled 2-1 for the Pentagon under the Federal Acquisition Supply Chain Security Act (41 U.S.C. § 4713): the DoD reasonably concluded that Anthropic’s ability to manipulate Claude through training created a supply chain risk, “bad motive” is not required, and the removal was for refusing the DoD’s contract terms, not for Anthropic’s advocacy. A separate designation under 10 U.S.C. § 3252 was set aside by the Northern District of California on Aug 27 2026, which found no “adversary” and no bad motive in Anthropic’s good-faith negotiation of usage limits, and remains under a permanent injunction. A law-firm bulletin of Oct 13 says no exclusion order under the D.C. Circuit’s statute had yet been issued, and contractors may still use Anthropic commercially but should keep it out of DoD work. The DoD confirmed on Oct 6 it had stopped using Anthropic products (BBC sources say Claude was still in Palantir’s Maven the week before). It is the clearest case of a lab’s acceptable-use policy colliding with a government customer.

Timeline

Every dated event in the wiki that involves Anthropic v. Department of Defense, newest first.

Connections

Anthropic v. Department of Defense and everything one link away. Drag to move, scroll to zoom, click a node to open it.Open in the full graph →

What links here

AI lab Anthropic

Government body US Department of Defense

Sources

  1. Taft, “D.C. Circuit Rules Pentagon Can Exclude Anthropic from Supply Chain”, Oct 13 2026