Overview
- U.S. District Judge Rita F. Lin told lawyers Thursday she saw no sufficient proof that Anthropic could disable or alter delivered models and took cross-motions for summary judgment under submission.
- The hearing followed the Pentagon’s March designation of Anthropic as a national security “supply chain risk” after contract talks stalled over the company’s refusal to permit use for mass domestic surveillance or autonomous lethal actions.
- Lin warned that the record suggests the label may have been punitive in response to Anthropic’s public criticism of the Defense Department, raising possible First Amendment and Administrative Procedure Act problems.
- Justice Department lawyers said agencies remain concerned about redlines, overseas operations, and handling of sensitive data, while the Pentagon told the court it plans to wind down Anthropic use by Sept. 30 and some agency pilots end Aug. 30.
- A ruling that narrows the government’s ability to apply supply-chain risk tags could change how federal procurement balances national security needs with companies’ safety limits and shape future AI contracting nationwide.