In a bold legal counteroffensive, AI developer Anthropic has filed sworn declarations disputing Pentagon claims that it poses an "unacceptable risk to national security." The Friday afternoon court submission pushes back aggressively against the government's assertions, which threatened to sever their partnership despite ongoing negotiations.

The court filings reveal a striking disconnect between public statements and private communications. Anthropic contends the government's case fundamentally misunderstands technical aspects of their AI technology and relies on objections never voiced during months of negotiations. This latest legal maneuver comes just one week after Trump reportedly terminated the relationship—contradicting Pentagon assurances that both parties were nearly aligned on key issues.