The Pentagon is currently testing AI models to identify the preferences of 25 of its top users, as it seeks alternatives to Anthropic’s Claude. This initiative is part of a broader effort by the Department of Defense to diversify its AI capabilities by evaluating multiple systems that can cater to various operational needs for defense operations.

Pentagon: The Pentagon is the headquarters of the United States Department of Defense and directs military strategy, operations, and technology adoption across the armed forces. In the current context, it is actively evaluating competing AI systems to reduce dependence on a single provider amid policy differences over model usage. This testing targets preferences among key internal users to support rapid integration of alternatives.
Anthropic: Anthropic develops frontier AI models with an emphasis on safety constraints and responsible deployment practices. Its Claude models have been used extensively by the Department of Defense for intelligence analysis, planning, and other mission-critical tasks. Recent disputes over usage restrictions have led the Pentagon to pursue other commercial options.

Model Testing: Evaluations are underway to determine which AI systems best match the requirements of power users within defense operations.
AI Diversification: The Department of Defense has started deploying models from multiple frontier providers to give warfighters varied tools suited to different operational needs.