Pentagon's blacklisting of Anthropic ruled unlawful by US judge
anthropic
| Source: HN | Original article
A U.S. judge has ruled that the Pentagon’s decision to blacklist AI firm Anthropic was unlawful, blocking the agency’s action.
A U.S. federal judge has ruled that the Pentagon’s decision to label AI start‑up Anthropic as a “supply‑chain risk” was unlawful. Judge Rita Lin of the Northern District of California found that the Trump administration violated the law when it singled out Anthropic after the company refused the Department of Defense’s request to permit its Claude model for mass domestic surveillance of Americans or for fully autonomous weapons. The ruling, issued on Thursday, affirms a preliminary injunction the court granted in March that halted enforcement of the blacklist while the case proceeds.
The decision matters because it curtails the Pentagon’s ability to unilaterally bar AI vendors from government contracts on vague security grounds. By deeming the blacklist retaliatory, the judgment reinforces procedural safeguards for private tech firms and signals that supply‑chain designations must meet clear statutory criteria. For Anthropic, the ruling removes an immediate barrier to doing business with the U.S. military, while for the defense establishment it raises questions about how to assess emerging AI risks without overstepping legal limits.
As we reported on 28 August, Anthropic had already secured a court win that blocked the blacklist; today’s judgment solidifies that outcome and clarifies the legal reasoning. The next steps will likely involve a Pentagon appeal, which could prompt a higher‑court review of the agency’s supply‑chain authority. Stakeholders will also watch for any policy revisions from the Department of Defense and for how the ruling influences other AI companies facing similar government scrutiny.
Sources
Back to AIPULSEN