What happened
- Federal Judge Rita Lin, in California, ruled on Thursday, August 27, that designating Anthropic a supply chain risk was unlawful.
- The ruling establishes that the decision by Defense Secretary Pete Hegseth constituted retaliation contrary to the First Amendment, in addition to being arbitrary and capricious, and that the company was denied the due process guaranteed by the Fifth Amendment.
- The designation, issued earlier this year, ordered federal agencies to stop working with the company. Anthropic had refused to relax safeguards that would have allowed its models to be used for fully autonomous weapons and mass surveillance of U.S. citizens.
- The judge noted inconsistencies in the government’s position: while the designation was in force, Hegseth separately proposed invoking the Defense Production Act with respect to the company, and the Department of Defense remained interested in contracting with it.
- Anthropic filed lawsuits in California and in Washington, D.C. The second one is still open.
Why it matters
- For any Chilean organization that hired a U.S. AI provider: continuity of service depended on a political decision that wasn’t in the contract and couldn’t be negotiated from here. The ruling confirms the decision was unlawful. It also confirms it was in force for months before anyone struck it down.
- Repairing a service outage through the courts takes longer than an operation can hold out. Anyone with a critical workflow resting on a single model, from a single provider, in a single jurisdiction, now knows how long that risk lasts when it materializes.
- Important for Chile: from December 1, Law 21.719 requires documenting the processing of personal data, including the processors involved. Being forced to switch AI providers stops being a technical problem and becomes a record-keeping problem before the Personal Data Protection Agency.
What’s next
- The parallel lawsuit in Washington, D.C., is still in progress. No announced date for a ruling.
- No timelines announced for the U.S. government’s next steps.
- December 1, 2026: Law 21.719 takes effect in Chile, 95 days from this note.
Bottom line
The same ruling notes that, while the designation was in force, the government kept working with Anthropic’s Mythos model on cybersecurity matters. During those months the company was, at the same time, a supply chain risk and a supplier.
Sources: Ruling of the District Court in California (Aug 27, 2026) · AI Insider (Aug 28, 2026)