OpenAI Signs Deal With Pentagon To Deploy Models On SIPRNet

OpenAI has reached an agreement with the U.S. Department of War to deploy its artificial intelligence models within the department’s classified network, according to statements attributed to OpenAI CEO Sam Altman and reports by Reuters and Bloomberg.

The deal centers on making OpenAI’s models available for use in a secure, classified environment operated by the Department of War. Details on the specific models to be deployed, the timeline for rollout, and the scope of initial use were not provided in the reports cited.

Altman said the agreement was reached “tonight,” referring to the timing of the announcement as carried by outlets including TradingView and republished by other publications. Reuters also reported the agreement, describing it as a deal to deploy AI models on the department’s classified network.

The Department of War is responsible for national security operations that handle highly sensitive information. Deploying AI tools inside classified networks is a significant step because it indicates the technology is being positioned for use in environments with strict access controls and high security requirements. It also reflects increasing interest across the federal government in using advanced AI systems for work that may involve protected data and mission-critical decision support.

For OpenAI, the agreement represents an expansion of its footprint into government and defense-related work beyond publicly accessible services and commercial enterprise products. For the Department of War, it signals a move toward bringing modern AI capabilities into secure systems where data cannot be exposed to the open internet or standard cloud environments.

The development comes amid heightened attention on which AI technologies federal agencies adopt and how those tools are governed. Recent headlines have included a report referenced by AOL.com about former President Donald Trump directing the government to cease using Anthropic’s technology after a Pentagon standoff. The OpenAI agreement arrives against that backdrop of scrutiny and competition among major AI providers for government contracts.

What happens next will depend on implementation steps that have not been publicly detailed in the available reports. Those steps typically include security reviews, integration work to connect models to approved systems, and setting parameters for how the tools can be used inside classified environments. Additional information could come from formal contract notices, agency statements, or further reporting that clarifies the operational scope of the deployment.

For now, the confirmed point is the agreement itself: OpenAI is set to place its AI models inside the Department of War’s classified network, marking a notable alignment between a leading AI developer and the U.S. national security apparatus.