Overview
- A U.S. district judge entered summary judgment for Anthropic on August 27, ruling the Pentagon’s February designation of the company as a “supply‑chain risk” violated the First and Fifth Amendments and enjoining the government from enforcing the challenged actions.
- The dispute began after Anthropic refused Pentagon demands to loosen ethical limits on a roughly $200 million classified AI contract, saying it would not allow its models for mass domestic surveillance or fully autonomous weapons.
- Judge Rita F. Lin said the record shows officials treated Anthropic’s public statements as a reason to punish the company and noted the government kept negotiating with Anthropic even after issuing the supply‑chain label.
- The ruling prevents the government from giving effect to the designation but does not compel the Pentagon to buy or use Anthropic’s models, and the administration is expected to appeal so final contracting outcomes remain unresolved.
- The case has large commercial and political stakes because Anthropic’s May funding round sharply raised its profile and valuation, it has drawn support from rival AI employees, and the decision could set limits on how national‑security tools are used against private vendors.