TL;DR
US federal judge Rita Lin has temporarily blocked the Trump administration from designating Anthropic a supply chain risk after the company refused to provide unrestricted military access to its AI. The ruling described the designation as “likely both contrary to law and arbitrary and capricious,” noting it could “cripple” the $380 billion (approximately £304 billion) company.
How negotiations broke down
Anthropic’s Claude model was the first AI system used by the US military in classified operations, reportedly including activities related to Iran and the capture of Venezuelan leader Maduro. But negotiations between Anthropic and the Pentagon collapsed over two specific demands: the use of AI in lethal autonomous weapons systems and mass surveillance applications.
When Anthropic refused, the Trump administration applied a supply chain risk designation — a tool designed to flag potential adversaries in government procurement. The designation had never previously been applied to an American company. Defence Secretary Pete Hegseth reportedly suggested that vendors end their commercial relationships with Anthropic entirely.
The court’s response
Judge Lin, sitting in the Northern District of California, did not hold back. She found the designation was likely unlawful and barred US agencies from implementing it for the time being. Her ruling included pointed language: “Nothing in the governing statute supports the Orwellian notion that an American company may be branded a potential adversary” for declining to remove safety restrictions from its products.
Several large technology companies filed supporting briefs on Anthropic’s behalf. The administration has seven days to appeal.
What this means for AI governance
This case establishes an early legal precedent for AI companies that draw ethical boundaries around their products. Until now, the question of whether an AI developer could refuse military applications without government retaliation was theoretical. It is now a matter of active case law.
The ruling also highlights a growing tension in US technology policy between the push for AI military superiority and the safety commitments that many AI companies have made publicly.
Looking forward
For UK AI companies with US government contracts or ambitions, this case is directly relevant. The UK’s own approach to military AI — outlined in the Ministry of Defence’s AI strategy — has been more measured, but UK firms operating in the US market face the same questions about where to draw ethical lines. The court’s willingness to protect a company’s right to set those boundaries will be closely watched across the Atlantic.