OpenAI Deploys Classified AI with DOW

This agreement materializes after the U.S. Department of War (DOW) severed ties with rival AI lab Anthropic. The schism occurred over Anthropic's refusal to remove ethical restrictions on its AI, which the Pentagon deemed was "jeopardizing critical military operations." President Trump ordered all federal agencies to cease using Anthropic's technology, with Defense Secretary Pete Hegseth designating the company a "supply chain risk." Anthropic, which had been the first to deploy its models on classified DoD networks, is now facing a six-month phaseout and plans to challenge the designation in court. The core of the dispute was Anthropic's contractual "red lines," which prohibited the use of its AI for mass domestic surveillance and in fully autonomous weapons systems. The company argued that current AI is not reliable enough for such applications and that surveillance would violate fundamental rights. OpenAI's deal with the Pentagon reportedly includes similar safeguards. CEO Sam Altman stated the agreement contractually prohibits using their AI for domestic mass surveillance and requires human responsibility for any use of force. The deployment will be cloud-only to prevent use on "edge devices" like drones, and OpenAI will have cleared engineers on-site to oversee integration and compliance. This layered approach to safety is what OpenAI claims makes their agreement more robust than previous classified AI deployments. The deal positions OpenAI as a key AI provider for the U.S. government, stepping in as its primary competitor is pushed out. It also highlights a significant moment in the relationship between AI labs and national security, demonstrating that negotiated safety protocols can bridge the gap between private industry ethics and government defense needs.