Following agreements with Google, SpaceX, and OpenAI, the U.S. Defense Department announced on Friday that it has reached agreements with Nvidia, Microsoft, Amazon Web Services, and Reflection AI, enabling it to utilize their AI technologies and models on its classified networks for “lawful operational use.”

“These partnerships expedite the transition to position the United States military as an AI-first combat force and will enhance our warfighters’ capacity to sustain decision superiority across all warfare domains,” the statement states.

These agreements emerge as the U.S. Department of Defense has hastened its diversification of AI suppliers following its controversial conflict with Anthropic regarding the usage terms of its AI models. The Pentagon sought unrestricted access to Anthropic’s AI resources, while the AI lab maintained that restrictions were necessary to prevent Anthropic’s technology from being employed for domestic mass surveillance and autonomous weaponry.

Currently, the two parties are engaged in legal battles, although Anthropic successfully obtained an injunction in March against the Pentagon’s attempt to label the company as a “supply-chain risk.”

“The Department will persist in developing an architecture that avoids AI vendor lock-in and guarantees long-term adaptability for the Joint Force,” the statement adds. “Access to a varied range of AI capabilities from the robust American technology stack will provide warfighters with the tools necessary to operate confidently and protect the nation against any threats.”

The DOD indicated that the companies’ AI hardware and models will be utilized in Impact Level 6 (IL6) and Impact Level 7 (IL7) settings to “facilitate data synthesis, enhance situational awareness, and support warfighter decision-making.” IL6 and IL7 represent high-security classifications for data and information systems identified as critical to national security, requiring stringent physical protections, access controls, and audits.

The Pentagon reported that over 1.3 million DOD personnel have utilized its secure enterprise platform for generative AI, GenAI.mil, which offers access to large language models (LLMs) and other AI tools within government-sanctioned cloud environments. It is primarily designed to assist with non-classified activities such as research, document preparation, and data analysis.