The US Department of War has begun offering a specialized version of OpenAI’s chatbot across its central generative-AI environment, expanding the tools available to millions of service members, civilians, and contractors for everyday unclassified work.
On August 31, 2026, officials activated ChatGPT Mil on GenAI.mil, the department’s enterprise platform.
The rollout represents the operational stage of a broader collaboration first formalized with OpenAI in 2025.
The customized system received accreditation to handle Controlled Unclassified Information at Impact Level 5, a security standard that permits use with sensitive but unclassified material while keeping it isolated from commercial versions of the same technology.
ChatGPT Mil is designed to feel familiar to anyone who has used the public product, yet it operates entirely inside the department’s protected infrastructure.
Users can converse with the model, upload and work with files, organize projects, and create custom GPTs tailored to specific tasks.
Additional capabilities are scheduled to appear in later phases.
Officials said the tool is especially useful for document-intensive processes that dominate much of the department’s administrative and planning workload—policy drafting, logistics coordination, contracting reviews, report generation, and similar functions.
The platform is built to serve the department’s full workforce of more than three million people.
By automating or accelerating routine writing, analysis, and synthesis, the system is intended to give personnel more time for higher-priority operational and decision-making work.
Its arrival also advances GenAI.mil from a single-model offering into a multi-vendor environment that already includes other frontier systems, giving users a choice of models depending on the task.
Oversight of the expansion sits with the Chief Digital and AI Office and the Office of the Under Secretary of War for Research and Engineering.
Department statements framed the launch as consistent with the White House AI Action Plan and with the department’s own strategy of treating advanced models as a force multiplier for readiness and decision advantage.
Data processed through ChatGPT Mil remains inside the government environment and is not used to train OpenAI’s public or commercial models.
Safeguards at both the model and platform layers are meant to support lawful uses while preserving human accountability for consequential decisions.
Officials have emphasized that the current deployment is limited to unclassified work; separate efforts are under way for classified networks and combat-related applications.
GenAI.mil itself launched in December 2025 and quickly attracted more than a million unique users.
The addition of ChatGPT Mil follows months of security testing and integration work that began after the partnership announcement earlier in the year.
The department presents the multi-model approach as a way to avoid dependence on any single vendor while giving warfighters practical tools that match the pace and volume of modern staff work.
The move illustrates how rapidly commercial generative-AI products are being adapted for large-scale government use once they meet defense-specific accreditation requirements.
For personnel already accustomed to the commercial ChatGPT interface, the military version lowers the barrier to using the technology on official documents and workflows that previously required more time-consuming manual effort.