Trump Admin Bans Anthropic, Taps OpenAI for Pentagon
The Trump administration has ordered all federal agencies and military contractors to cease using Anthropic's AI products, designating the company a “supply chain risk” amid a dispute over AI safety controls for military use. In a dramatic policy shift, the Pentagon is now transitioning to OpenAI, which promptly struck a new deal for defense-related AI systems.
The core of the dispute was Anthropic's refusal to grant the Pentagon "unrestricted use" of its AI models, specifically objecting to their potential application for mass surveillance and in fully autonomous weapons systems without meaningful human oversight. Anthropic's CEO, Dario Amodei, stated the company could not "in good conscience accede" to these demands, arguing current models are not reliable enough to avoid lethal mistakes. In an unprecedented move against a U.S. company, Defense Secretary Pete Hegseth designated Anthropic a "supply chain risk," a label historically reserved for foreign adversaries like China's Huawei. This designation immediately bars any contractor, supplier, or partner doing business with the U.S. military from conducting any commercial activity with Anthropic, effectively blacklisting it from the defense ecosystem. The Pentagon's pivot to OpenAI is part of a broader, pre-existing relationship under the "OpenAI for Government" initiative, which already included a $200 million contract to develop prototype AI for national security. OpenAI CEO Sam Altman confirmed their new agreement includes building technical safeguards and incorporates the company's principles prohibiting mass surveillance and requiring human responsibility for the use of force. This conflict is a critical case study in the governance of agentic AI, where systems are designed not just to respond, but to act autonomously to achieve goals. The Pentagon's push for unrestricted use in military reasoning and planning highlights the tension between the capabilities of these autonomous workflow patterns and the ethical guardrails established by their creators. The fallout disrupts significant ongoing operations, as Anthropic's Claude was the first frontier AI model deployed on classified U.S. government networks, actively used by agencies like the CIA and NSA for intelligence analysis. President Trump has mandated a six-month phase-out period for the Pentagon to transition away from Anthropic's embedded technology. The ban creates a vacuum that competitors are positioned to fill, with reports suggesting Elon Musk's xAI and its Grok model may be granted access to classified networks. Musk publicly supported the administration's decision, illustrating the intense geopolitical and competitive dynamics among frontier AI labs vying for dominance in the defense sector.
