The Pentagon Now Has Its Own Version of ChatGPT and Grok

My take: The Department of Defense went from 80,000 AI users to 1.7 million in under nine months. That is not a pilot program: that is institutional adoption at real scale. With the addition of ChatGPT Mil (OpenAI) and Grok for Government (Starshield/SpaceX) to the GenAI.mil platform, the Pentagon now offers three frontier models to its 3 million employees and service members.

What matters here is not which model is best. It is that the most demanding institution in the U.S. government just validated AI as a daily work tool for millions of people. If the Pentagon has already made that call, the debate over whether to use AI at work should be settled for any business.

For those of us building with Claude: its absence from the platform is not a technical issue. It is a foreign policy issue, and a reminder that vendor strategy matters. Does your company or team have clarity on who you depend on for your AI tools?