Pentagon adds ChatGPT and Grok to its AI portal

The Department of War added OpenAI’s ChatGPT Mil and xAI’s Grok for Government to its GenAI.mil platform on August 31, joining Google’s Gemini, which had been the portal’s only option since launch. GenAI.mil now covers roughly 3 million military members, civilian employees, and contractors, with more than 1.7 million already using the platform. Both new products cleared Impact Level 5, the Pentagon’s highest authorization tier for non-public, sensitive unclassified data.

ChatGPT Mil runs on GPT-5.4 Terra today with GPT-5.6 Terra coming, and targets document-heavy unclassified work: administrative tasks, logistics, planning, and policy drafting, plus offline search with citations and automation for repetitive paperwork like travel forms and contracting responses. Grok for Government adds adaptive reasoning modes (Auto, Fast, Expert), persistent project workspaces, and reusable “playbooks” meant to preserve institutional knowledge between staff turnover. A War Department official framed the expansion as giving “warfighters tools to act with confidence.”

Anthropic’s Claude is the notable absence from a three-vendor field that could have been four. Claude was excluded after Anthropic refused to loosen the restrictions it holds on mass-surveillance and lethal-autonomous-weapons use, a dispute that led the Pentagon to label the company a supply-chain risk before a federal judge ruled that designation unlawful retaliation and vacated it. That ruling hasn’t yet translated into a GenAI.mil listing for Claude. The expansion is the latest sign of how much defense budget is now flowing toward AI vendors generally, alongside deals like Helsing’s $18 billion push into defense AI.