A federal judge in California has ruled that the Pentagon’s decision to blacklist Anthropic earlier this year was unconstitutional. The ruling concludes a legal dispute where the Trump administration allegedly retaliated against the AI developer for establishing safety restrictions on military applications of its technology. District Judge Rita F. Lin determined that citing national security concerns was not a valid excuse to punish a government contractor simply for setting ethical boundaries. The court noted that empty invocations of security cannot serve as a blank check to retaliate against critics. This decision validates the company’s claim that the blacklisting violated the First Amendment and the Administrative Procedure Act. The order effectively restores Anthropic’s ability to supply its models to the US defence sector, ending a period of uncertainty regarding its commercial contracts.
The outcome clarifies the legal limits executive agencies face when restricting access to advanced computing tools. It establishes that safety protocols do not constitute grounds for punishing private firms under national security pretexts. This sets a precedent for future disputes involving government procurement and AI regulation.
- The ruling was issued in the Northern District of California.
- Anthropic had filed the lawsuit in March.
- The judge cited violations of the First Amendment.
