OpenAI Inks Pentagon Deal
Just hours after the Trump administration banned rival Anthropic as a “supply chain risk,” OpenAI secured a major contract with the Pentagon. The deal gives the Defense Department access to OpenAI's advanced AI models, with CEO Sam Altman announcing it includes “technical safeguards” for responsible deployment. The move cements OpenAI's position as the U.S. government's preferred AI vendor and signals a new era of geopolitical considerations in AI procurement.
The deal followed a contentious public dispute between the Pentagon and rival AI firm Anthropic. The Trump administration designated Anthropic a "supply chain risk" and ordered federal agencies to cease using its technology after the company refused to remove contractual safeguards against the use of its AI for mass surveillance and fully autonomous weapons. Anthropic has stated it will challenge the designation in court. OpenAI's agreement includes similar "red lines," explicitly prohibiting the use of its models for domestic mass surveillance and requiring human responsibility for the use of force. CEO Sam Altman stated these principles are reflected in law and policy and were integrated into their agreement, a point the Pentagon reportedly accepted after refusing Anthropic's similar requests. As part of the deal, OpenAI's models will be deployed exclusively in the cloud, not on "edge" devices like drones, to prevent direct integration with weapons systems. The company will also retain control over its "safety stack" and embed its own cleared personnel to monitor the AI's use within the Pentagon's classified networks. This single contract is a fraction of the Pentagon's larger investment in artificial intelligence. The Department of Defense's fiscal year 2026 budget request includes a record $13.4 billion for AI and autonomy, signaling a major strategic push to integrate these technologies across military operations. The agreement has drawn mixed reactions. Top House Democrat Hakeem Jeffries praised Anthropic's "courage" for pushing back against what he called a "shocking invasion of privacy scheme." Meanwhile, a labor coalition representing over 700,000 tech workers has urged major tech companies to follow Anthropic's lead in resisting military contracts that require stripping safety protocols. The Pentagon's Chief Digital and AI Office (CDAO) is overseeing the integration, which is part of a broader effort to partner with multiple commercial AI leaders, including Google and xAI. The initial phase involves prototyping AI for administrative tasks, such as improving healthcare for service members, streamlining procurement data analysis, and bolstering cyber defense. The full integration will be phased, starting with testing in a secure "sandbox" environment, followed by limited field deployments by the end of 2025. A full operational capability of the AI models within the Department of Defense is targeted for the second half of 2026. The move is seen by some as setting a precedent for how the U.S. military procures and governs advanced AI, potentially creating a template for future defense contracts that balance access to cutting-edge technology with ethical safeguards. However, some critics view OpenAI's deal as an opportunistic move that undermines the ethical stand taken by Anthropic.
