US Military's Secret Weapon: The Pentagon Builds Its Own ChatGPT and Grok
The Pentagon has launched custom generative AI tools, ChatGPT Mil and Grok for Government, through its secure GenAI.mil portal, granting 1.7 million personnel access. This initiative aims to boost productivity and secure a technological edge while addressing past disputes with companies like Anthropic over AI use.
The Pentagon has significantly expanded its embrace of generative artificial intelligence, providing 3 million civilian and military personnel with access to customized versions of OpenAI’s ChatGPT and xAI’s Grok. These tailored AI tools, branded as ChatGPT Mil and Grok for Government, are now integral components of GenAI.mil, a secure, centralized portal launched last year. This initiative aims to equip Department of Defense employees with commercial frontier AI models while safeguarding sensitive government data from ordinary consumer channels and their associated data collection practices.
GenAI.mil, which initially offered Google Gemini, has quickly gained traction within the department. According to the Department of Defense, the portal has already onboarded more than 1.7 million unique users, demonstrating a substantial adoption rate among its 3 million personnel. This widespread deployment underscores the Pentagon’s broader strategy to leverage AI tools for accelerating work processes, enhancing productivity, and securing a technological advantage without compromising security.
Notably, these recent additions highlight the absence of Anthropic’s Claude model from the GenAI.mil platform. This omission stems from an earlier dispute where Anthropic, designated a supply-chain risk by the Trump administration (a decision currently under legal challenge), refused to grant the Pentagon unrestricted use of its AI tools, insisting instead on specific safety guardrails. Consequently, the Pentagon has diversified its partnerships, collaborating with other companies in its pursuit of advanced AI capabilities.
ChatGPT Mil, developed under OpenAI’s Government program, is designed to offer an experience familiar to commercial ChatGPT users. The tool will concentrate on core functionalities such as chat, file management, project collaboration, and custom GPTs. Its primary application will be in handling document-heavy, routine unclassified tasks, including administrative duties, logistics, planning, and policy development, with additional features slated for future integration.
In contrast, the Pentagon has framed Grok for Government’s capabilities in broader, more militaristic terms. According to the Defense Department, Starshield AI’s Grok for Government is expected to deliver “immediate productivity gains, stronger knowledge continuity and more secure and efficient collaboration” for warfighters. Powered by SpaceX’s secure satellite network, Starshield AI, Grok is envisioned to empower military personnel to “execute missions faster and with greater precision across numerous operational contexts,” ranging from market research analysis for acquisition professionals to supply-chain management for logisticians.
Beyond these primary generative AI tools, the Pentagon has forged strategic deals with a variety of other leading technology companies to further bolster its AI capabilities. These partners include major players like Amazon Web Services, Microsoft, Nvidia, and Reflection AI, indicating a comprehensive effort to integrate advanced AI across various departmental functions.