OpenAI lands Pentagon deal to deploy AI, Anthropic sidelined in safety clash

OpenAI CEO Sam Altman announced a deal with the Department of War to deploy OpenAI models in the agency's classified networks, following Anthropic's withdrawal from the arrangement amid a standoff with the Pentagon. Altman said, 'Tonight, we reached an agreement with the Department of War to deploy our models in their classified network. In all of our interactions, the DoW displayed a deep respect for safety and a desire to partner to achieve the best possible outcome.' The Department of War designated Anthropic a 'supply chain risk' after it failed to reach an agreement with the agency, a move that coincided with President Trump ordering federal agencies to stop using Anthropic.

Under the deal, OpenAI and the Pentagon have agreed on two core safety principles: they prohibit domestic mass surveillance and require human accountability in decisions involving the use of force, including autonomous weapons systems. The Department of War supports these principles, embeds them in law and policy, and has incorporated them into the agreement. Altman stated that the DoW's implementation of these terms is a model for the industry and that OpenAI would like to see the same terms offered to all AI companies, in hopes of de-escalating legal and governmental actions and moving toward reasonable agreements.

The arrangement underscores a high-stakes balancing act between national security interests and the push for rapid AI deployment. Anthropic's exit highlights the friction over how AI should be governed in sensitive government contexts. The broader implications are clear: as federal agencies increasingly explore advanced AI capabilities, safety and accountability provisions may become a frontline requirement for any future partnerships, potentially shaping how the tech industry negotiates with government buyers.

#OpenAI #PentagonAI #AISafety