Pentagon Escalates Pressure on Anthropic AI
Defense Secretary Hegseth has reportedly warned AI firm Anthropic to relax its model safeguards for military applications or risk being blacklisted from future contracts. The warning marks an escalation in the dispute over responsible AI principles versus defense requirements. The tensions were highlighted by a viral hoax, shared by investor Bill Ackman, that falsely claimed Anthropic had already capitulated to DoD pressure.
- The confrontation is a direct result of Defense Secretary Hegseth's "AI Acceleration Strategy," unveiled in January 2026, which commits the Pentagon to becoming an "AI-first warfighting force." Hegseth has publicly stated the DoD "will not employ AI models that won't allow you to fight wars" and that AI must be "without ideological constraints that limit lawful military applications." - The Pentagon has threatened to invoke the Defense Production Act, a law typically used for industrial mobilization during national emergencies, to compel Anthropic to remove its safeguards—a potentially unprecedented use of the act. Another threat is designating Anthropic a "supply chain risk," which would prohibit all DoD contractors from using its technology. - The dispute reportedly escalated after Anthropic's Claude AI was used through partner Palantir'
