Pentagon Signs Classified AI Deals with Major Tech Firms, Excludes Anthropic

The Pentagon on Friday signed classified AI deployment agreements with seven major tech firms — SpaceX, OpenAI, Google, Nvidia, Reflection, Microsoft, and Amazon Web Services — integrating their models into sensitive military networks for mission planning and weapons targeting. The move deliberately excludes Anthropic, whose Claude model had been the only one authorized for classified operations until a dispute erupted over the company's refusal to allow its AI for mass surveillance or direct control of lethal autonomous weapons. The Trump administration labeled Anthropic a supply-chain risk in February and ordered a halt to its use.

Emil Michael, the Pentagon's de facto CTO, emphasized the need to avoid reliance on any single vendor. The deals include both closed and open-source models, with Nvidia and Reflection providing open-source options for greater operational flexibility. The Pentagon's GenAI.mil platform has already been used by 1.3 million personnel, generating tens of millions of prompts and deploying hundreds of thousands of AI agents in five months. Meanwhile, over 600 Google employees demanded the company reject the Pentagon deal, echoing the 2018 Project Maven protests that forced Google to abandon a drone AI program.