Accredited at Impact Level 5
The US Department of War said it has launched OpenAI's ChatGPT Mil on GenAI.mil, its department-wide generative AI platform. The build is accredited for controlled unclassified information at Impact Level 5 and is meant to serve more than 3 million personnel.[1]
The initial release covers chat, files, projects and custom GPTs, with more features described as coming later. The launch follows an enterprise agreement with OpenAI signed in 2025, and the department ties it to the White House AI action plan.[1]
Planning, policy, logistics and administration
The department presents the launch as part of a multi-model platform and names document-heavy unclassified work — planning, policy, logistics and administration — as the intended use. ChatGPT Mil is meant to serve more than 3 million personnel on that Impact Level 5 work.[1]
Oversight has named offices
Oversight sits with the Chief Digital and AI Office and the Office of the Under Secretary of War for Research and Engineering. The more than 3 million personnel figure is the population ChatGPT Mil is meant to serve. Planning, policy, logistics and administration sit in a different quantity: time on those tasks. An alternative is that the Impact Level 5 build may add a channel for controlled unclassified information while those tasks stay with the same personnel.[1]
The next measurable line is time spent on planning, policy, logistics and administration, or time given to oversight of that work, published beside ChatGPT Mil use. The offices that hold oversight are already named for the more than 3 million personnel the tool is meant to serve.[1]