OpenAI has officially joined the Pentagon's list of suppliers after signing a $200 million contract with the US Department of Defense . The collaboration concerns the development of advanced artificial intelligence (AI) systems, with an emphasis on applications related to proactive cyber defense and optimization of internal military operations.
According to an official announcement from the Pentagon, OpenAI is undertaking the development of original and advanced AI technologies aimed at addressing critical national security challenges — both at the military and administrative levels. The project is planned to be carried out primarily in the Washington region and is expected to be completed by July 2026.
See also: OpenAI: ChatGPT Memory upgrade for “free users”
OpenAI confirmed the partnership in a blog post, describing it as the first agreement in a new strategy to provide AI technologies to federal, state, and local government agencies. The company said the models offered for national security are tailored and subject to clear usage policiesthat exclude their application in offensive military operations or actions that cause harm to people or property.
“This contract, capped at $200 million, will leverage OpenAI’s industry-leading expertise to help the Department of Defense understand how cutting-edge AI can transform its administrative functions, from improving how service members and their families receive healthcare, to optimizing how they view program and procurement data, and supporting proactive cyber defense,” the company noted.
See also: OpenAI plans to combine multiple models in GPT-5
While this contract officially marks OpenAI's entry into the national security sector US, it's not the first time the company has worked with the military. In December 2024, OpenAI began a partnership with Anduril Industries, integrating its technologies into drones .
It is worth noting that the new agreement comes in stark contrast to OpenAI's previous policies, which prohibited the use of its models for "military or war purposes" — a policy that was repealed last year, paving the way for the current partnership.
The contract is part of a broader shift by the US government towards adopting AI in the defense sector, with similar moves by other technology companies. Anthropic recently launched an AI model with fewer restrictions specifically for military use, while Google and Meta revised their commitments on the “offensive” use of AI, now allowing its application in national security.
See also: OpenAI explains which ChatGPT model you need
From a technological perspective, this shift is an expected development. AI systems can significantly transform the operation of military and administrative infrastructure — from logistics and cybersecurity to big data analysis. OpenAI has the expertise to support such projects, especially at the level of natural language modeling, threat prediction, and proactive intelligence.
However, ethically and politically, such contracts raise concerns. OpenAI was founded with the goal of “safe and beneficialAI for all.” Working with the military complex, especially in a world of heightened geopolitical tensions, raises questions about whether the use of AI can be kept within peaceful contexts — even when the intent is “defensive.” Also, OpenAI’s removal of old restrictions against use for war purposes suggests a shift in its strategy, possibly under pressure or due to increased competition in the field.
Source: www.theverge.com
