OpenAI Deploys on Classified Networks
OpenAI has reached a new, classified deployment agreement with DOW, allowing its AI models to be used on secure government networks. The deal marks a significant step in integrating advanced commercial AI into sensitive federal applications, moving beyond the previously announced Pentagon partnership.
The deal materialized swiftly after rival AI firm Anthropic refused to eliminate safeguards preventing its technology from being used for autonomous weapons and domestic surveillance. The Pentagon had given Anthropic a deadline to allow for "all lawful purposes," which the company, led by CEO Dario Amodei, stated it could not "in good conscience accede" to. This public dispute culminated in Defense Secretary Pete Hegseth designating Anthropic a "supply chain risk" and President Donald Trump ordering all federal agencies to cease using its technology. Within hours of the breakdown in negotiations, OpenAI, led by CEO Sam Altman, announced it had reached an agreement to deploy its models on the Pentagon's classified systems. OpenAI's agreement includes specific "red lines" that mirror the safeguards Anthropic had advocated for, prohibiting the use of its AI for mass domestic surveillance or to direct autonomous weapons systems without human control. Following public criticism that the initial deal was "opportunistic and sloppy," OpenAI amended the contract to explicitly forbid its models from being used by intelligence agencies like the NSA and for domestic surveillance using commercially acquired data. This partnership is part of a broader push by the Pentagon to integrate advanced AI, a strategy that predates this specific agreement. A key predecessor is Project Maven, initiated in 2017 to use AI for analyzing drone footage and improving targeting. The Department of Defense has also launched GenAI.mil, an unclassified platform that has already attracted over a million users across military branches, signaling a wide-scale adoption of generative AI. The OpenAI contract, valued at up to $200 million, will focus on developing prototype AI for both "warfighting and enterprise domains," including improving healthcare for service members and cyber defenses. This places OpenAI alongside other major tech players like Google, Palantir, and Anduril, which also have significant contracts to advance the Pentagon's AI capabilities. Cleared OpenAI engineers will be deployed to assist the government with the new systems, which will be restricted to cloud-based environments rather than on "edge systems" like drones. The move signifies a deeper integration of commercial AI into sensitive national security operations, moving beyond administrative tasks to more critical applications.
