OpenAI has secured a substantial contract worth $200 million from the United States Department of Defense (DoD) to deliver advanced artificial intelligence (AI) solutions.
The agreement, spanning one year, marks OpenAI’s first officially documented partnership with the DoD and shows the increasing integration of cutting-edge AI technologies within national security and defence operations.
The announcement follows OpenAI’s earlier collaboration with Anduril Industries, a notable defence technology startup, that aims to deploy sophisticated AI systems for national security missions.
Anduril itself was awarded a $100 million defence contract in late 2024. At the same time, OpenAI’s competitor Anthropic has been forging partnerships with Palantir and Amazon to supply AI models tailored for U.S. defence and intelligence agencies.
The DoD has tasked OpenAI with developing prototype “frontier AI capabilities” to tackle critical challenges in both combat and administrative domains. This initiative is part of OpenAI’s programme, OpenAI for Government, which includes the existing ChatGPT Gov product. The programme intends to provide U.S. government bodies with customised AI models, along with ongoing technical support and product development.
OpenAI has emphasised that the contract will explore how AI can transform various administrative functions within the military, such as improving healthcare delivery for service members and their families, streamlining the analysis of programme and acquisition data, and boosting proactive cyber defence capabilities. The company has committed to ensuring all applications comply strictly with its usage policies and ethical guidelines.
Sam Altman, OpenAI’s CEO and co-founder, expressed the company’s commitment to national security efforts during a discussion at Vanderbilt University in April 2025. Speaking alongside Paul Nakasone, a former National Security Agency director and current OpenAI board member, Altman said, “we have to and are proud to and really want to engage in national security areas.”
The contract work will predominantly take place within the National Capital Region, which includes Washington, D.C., and surrounding counties in Maryland and Virginia, aligning with the concentration of U.S. defence and intelligence agencies.
In addition to this contract, OpenAI is actively expanding its AI infrastructure within the United States. Earlier this year, Altman appeared alongside President Donald Trump at the White House to announce the ambitious Stargate project, a $500 billion initiative designed to build robust AI infrastructure domestically.
While the $200 million contract constitutes a modest portion of OpenAI’s overall revenue, it represents a notable step in the company’s engagement with defence. OpenAI’s annualised revenue has surpassed $10 billion, and the company’s valuation reached approximately $300 billion following a $40 billion funding round announced in early 2025, showing strong investor confidence.
Microsoft, which provides cloud infrastructure to OpenAI, has also advanced government AI applications. In April 2025, the U.S. Defense Information Systems Agency authorised the use of Microsoft’s Azure OpenAI service for handling classified information at the secret level. This authorisation enables the secure deployment of AI tools within sensitive government environments, further facilitating AI integration into defence operations.