Judge questions Pentagon's labeling of Anthropic as supply-chain risk in ongoing legal dispute

Federal Judge Rita Lin of the Northern California district court questioned on March 25, 2026 whether the Pentagon is attempting to 'cripple' Anthropic after the Trump administration designated the AI company as a supply-chain risk for advocating increased regulation of AI-powered weapons. The designation, which would block Anthropic from military contracts and government contractor work, makes Anthropic the first US company to receive such a label from the Defense Department. The case has attracted widespread support, with Microsoft, OpenAI employees, Google employees, and Catholic ethicists filing court briefs backing Anthropic's stance on requiring human oversight for AI weapons due to reliability concerns like hallucinations. The dispute occurs amid broader political battles over AI regulation, with competing super PACs spending over $100 million on congressional races, and as Anthropic's Claude Gov models are reportedly being used in US-Israel combat operations against Iran for target selection.

Pentagon's designation of Anthropic as supply-chain risk and legal challenge

• ▪Employees of OpenAI and Google Inc filed court briefs in support of Anthropic's stance on AI regulation
• ▪Judge Rita Lin of the Northern California district court stated on March 25, 2026 that the Pentagon's designation of Anthropic looks like an attempt to cripple Anthropic
• ▪Anthropic is the first US company to be designated as a supply chain risk by the Defense Department
• ▪On March 17, 2026, the Defense Department told the court that Anthropic's stance on AI-powered weapons would undercut the Defense Department's ability to control its own lawful operations
• ▪Judge Rita Lin stated that the United States Department of Defense may be illegally trying to punish Anthropic for attempting to restrict the use of Anthropic's AI models for weapons without human supervision or for mass surveillance
• ▪The Trump administration designated Anthropic as a supply chain risk for Anthropic's stance on increased regulation of AI-powered weapons
• ▪Catholic Moral Theologians and Ethicist filed court briefs in support of Anthropic's stance on AI regulation

Technical limitations and safety concerns of AI in military applications

• ▪OpenAI and Google researchers stated in court submissions that more than 70 million cameras and credit card transaction histories can be collated to monitor the entire US population
• ▪Mary Cummings found that half of all accidents by self-driving cars in San Francisco were caused by phantom braking from AI hallucination
• ▪OpenAI and Google DeepMind engineers stated in court filings that AI decisions in lethal contexts are irreversible
• ▪Anthropic asked for continued human oversight in Anthropic's contract with the Defense Department because AI models can hallucinate and are not yet completely reliable
• ▪Engineers from OpenAI and Google DeepMind stated in court filings that AI models' chain of reasoning is often hidden from operators and internal workings are opaque even to developers
• ▪Anthropic's Claude Gov models have been integrated into Palantir's Project Maven for data analysis and target selection
• ▪Mary Cummings warned in a February 2026 paper that the incorporation of AI into weapons will face similar reliability issues as self-driving cars, including hallucinations
• ▪Brianna Rosen stated that the United States is using AI to generate targets in large-scale combat operations in Iran for the first time

Broader debate over AI regulation and political influence

• ▪Leading The Future super PAC has funded advertisements against Alex Bores, a New York assembly member running for Congress
• ▪In February 2026, Anthropic announced a $20 million donation to Public First Action, a PAC that will support candidates in favor of AI regulation
• ▪A Quinnipiac University poll in April 2025 found that 69 percent of Americans thought the government could do more to regulate AI
• ▪Leading The Future super PAC has received more than $100 million from OpenAI President Greg Brockman and Palantir co-founder Joe Lonsdale
• ▪Public First Action will support Alex Bores's congressional campaign