Pentagon Pressures Anthropic on Military Use as $200 Million AI Deal Hangs in Balance

New updates have been reported about Anthropic.

Anthropic is resisting a Pentagon push to allow its Claude AI models to be used by the U.S. military for “all lawful purposes,” putting a reported $200 million defense contract at risk, according to Axios. The company has emerged as the most reluctant among major AI vendors facing similar demands, as Defense Department officials seek broad usage rights that could include sensitive operational contexts.

The dispute centers on Anthropic’s strict usage policies, with a spokesperson telling Axios the firm is not negotiating over specific military operations but is focused on preserving hard limits on fully autonomous weapons and mass domestic surveillance. Tensions follow earlier reports of internal disagreements over how Claude could be deployed by the Pentagon, and claims that the model was used in a U.S. operation targeting then–Venezuelan President Nicolás Maduro, underscoring the strategic and reputational stakes for Anthropic as it balances lucrative government work against governance commitments that may constrain future defense-sector revenue and influence procurement decisions.