The Pentagon reportedly threatened to end its $200 million contract with Anthropic after the AI firm refused to allow its tools to be used in autonomous weapons and mass surveillance.
The US military wants AI labs to let it use their models “for all lawful purposes,” Axios reported, and apparently used Anthropic’s Claude in the operation to capture Venezuelan President Nicolás Maduro. AI is increasingly used in defense, and major commercial suppliers permit its use in unclassified work, such as generating reports and assisting workflow.
Its use in actual warfighting is controversial but likely inevitable: AI-powered drone swarms are nearing battlefield use in Ukraine, while human-piloted fighter jets will soon have unmanned, AI-controlled “loyal wingmen” flying alongside.