The US Pentagon has launched custom versions of OpenAI’s ChatGPT and xAI’s Grok, giving millions of military and civilian personnel access to generative artificial intelligence tools tailored to defence needs.
The new services, ChatGPT Mil and Grok for Government, have been added to GenAI.mil, a centralised and secure AI platform designed to provide US Department of Defense personnel with access to commercial AI models.
The Pentagon said the platform is intended to allow employees to use advanced AI systems without sending sensitive government information through ordinary consumer versions of such services.
More than 1.7 million Pentagon personnel using AI portal
According to the US Defence Department, more than 1.7 million unique users have already accessed GenAI.mil out of approximately 3 million military and civilian personnel.
The platform initially offered access to Google’s Gemini and is now expanding its selection of AI systems as the Pentagon seeks to accelerate administrative and operational work while maintaining security.
ChatGPT Mil, developed through OpenAI’s government programme, will offer features familiar to commercial ChatGPT users, including chat, file handling, projects and custom GPTs.
The Pentagon said the system can support routine, document-heavy unclassified work, including administration, logistics, planning and policy-related tasks.
Grok positioned for broader military use
The Pentagon has described Grok for Government in broader operational terms, saying the system could help military personnel improve productivity, knowledge sharing and collaboration.
According to the department, Grok could be used for tasks ranging from market research and acquisition analysis to supply-chain management.
The launch is part of the Pentagon’s wider effort to integrate advanced AI into its operations while keeping government data within secure systems.
Anthropic’s Claude absent from Pentagon portal
The expansion also highlights the Pentagon’s strained relationship with Anthropic, whose Claude AI model is not included among the latest additions.
The US administration previously designated Anthropic a supply-chain risk after the company resisted demands for unrestricted military use of its AI systems and insisted on maintaining certain safety safeguards.
The Pentagon has instead expanded partnerships with other technology companies, including Amazon Web Services, Microsoft, Nvidia and Reflection AI, as it seeks to strengthen its AI capabilities.
The latest rollout reflects the growing role of generative AI in US defence operations, with the Pentagon seeking to balance faster adoption of advanced technology with security and operational safeguards.
