It’s official: The Pentagon has labeled Anthropic a supply-chain risk

The Department of Defense has officially designated Anthropic and its Claude AI models as a supply-chain risk, an unprecedented move typically reserved for foreign adversaries. The designation stems from a direct conflict with CEO Dario Amodei, who refused Pentagon requests to use Anthropic's technology for domestic mass surveillance of Americans and for powering fully autonomous weapons systems without human oversight in targeting or firing decisions. The Pentagon argued its AI use should not be limited by a private contractor, leading to the retaliatory designation that now requires any Pentagon contractor to certify they don't use Anthropic's systems. This threatens to disrupt U.S. military operations, as Claude has been the only frontier AI lab with classified-ready systems and is actively deployed in Palantir's Maven Smart System for managing data in the Iran campaign.

The designation has sparked significant backlash, with hundreds of employees from OpenAI and Google urging Congress to intervene and calling the action an inappropriate use of authority against a domestic innovator. Critics, including former Trump White House AI adviser Dean Ball, have labeled it a 'death rattle' of American republic values. Meanwhile, OpenAI has forged a separate deal with the DOD allowing military use for 'all lawful purposes'—language that concerns some employees about potential surveillance and autonomous weapons applications. The conflict highlights a fundamental divide in the AI industry between companies willing to accommodate military demands and those, like Anthropic, establishing ethical red lines that prioritize human oversight and privacy protections.