Anthropic’s Pentagon Talks Hit Surveillance and Weapons Snag

Anthropic’s contract renewal talks with the Pentagon have stalled over disputes about AI usage restrictions for its Claude tool, according to an insider. The AI firm demands safeguards to block Claude from enabling mass surveillance of U.S. citizens or autonomous weapons development without human oversight. The Pentagon opposes these limits, arguing Claude should be usable as long as deployments remain legally compliant, per private negotiation details. Earlier reports by Axios first highlighted the impasse, framing it as a clash between ethical AI constraints and military operational flexibility. The deadlock underscores growing tensions between tech companies enforcing AI ethics and government agencies prioritizing unrestricted access to advanced tools.