Anthropic, OpenAI, Google, and xAI have each secured contracts worth up to $200 million from the US Department of Defense. The goal: bring advanced generative AI models into military and strategic operations.
The deals, awarded by the Pentagon's Chief Digital and Artificial Intelligence Office (CDAO), are designed to accelerate the adoption of cutting-edge generative AI - often called "frontier AI" - for national security. Over the next two years, the four companies will work with the Pentagon to explore how generative AI can be used across military, intelligence, and administrative settings.
Frontier AI for Military Workflows
The contracts use "Prototype Other Transaction Agreements" (POTA), a flexible contracting approach outside standard federal procurement rules. This structure is meant to help the Pentagon move technological innovations quickly into real-world operations.
The companies are tasked with developing AI agent prototypes that can plug directly into military and organizational workflows. Priorities include automating analysis, supporting decision-making with large-scale data, and integrating AI with existing information systems.
The Pentagon is taking a "commercial-first" approach: it wants to embed the best available commercial AI technology directly into its own systems.
Generative AI Access Across the Department
Beyond these projects, the CDAO is also making generative AI models available department-wide. These models are being integrated into existing DoD platforms like Ask Sage, Advana, Maven Smart System, and Edge Data Mesh, which already support a range of defense activities.
According to CDAO chief Doug Matty, the move is meant to give the US an edge over geopolitical rivals.
