OpenAI Inks Pentagon Deal
OpenAI has secured a landmark agreement to provide the Pentagon with its advanced AI models. The deal comes just after the U.S. government blacklisted rival AI firm Anthropic, citing "supply chain risks." The rapid pivot embeds OpenAI's technology directly into U.S. national security and defense planning.
The deal, valued at up to $200 million, is for the development of "prototype frontier AI" for both administrative and operational military purposes, with an expected completion date of July 2026. The project is being managed by the Pentagon's Chief Digital and Artificial Intelligence Office (CDAO). This marks a significant policy shift for OpenAI, which until early 2024 had a public policy that banned military-related work. OpenAI CEO Sam Altman stated the agreement includes firm "red lines," prohibiting the use of its AI for mass domestic surveillance, integration into autonomous weapons systems, and high-impact automated decisions like social credit scoring. The company's models will be deployed exclusively on secure cloud networks, not on edge devices like drones, and OpenAI will retain full control over its safety and classifier systems. The agreement materialized after talks between the Pentagon and rival AI firm Anthropic collapsed. The dispute centered on the Pentagon's demand for "all lawful use" of Anthropic's models, a clause Anthropic rejected, citing concerns about potential legal loopholes related to autonomous weapons and surveillance. In response to Anthropic's refusal to bend on its safety standards, Defense Secretary Pete Hegseth threatened to blacklist the company. Subsequently, the Trump administration officially designated Anthropic a "supply chain risk," a label historically used for foreign adversaries, and ordered all federal agencies to cease using its technology.
