OpenAI Robotics Chief Resigns Over Pentagon AI Deal

OpenAI reached an agreement with the U.S. Defense Department on February 27, 2026, allowing its AI models to be deployed in the Pentagon's classified cloud environment. The value of the partnership was not disclosed. The move came shortly after Anthropic faced losing a $200 million contract because it refused to loosen restrictions on mass domestic surveillance and fully autonomous weapons, underscoring governance and accountability risks as generative AI enters defense decision-making.

Caitlin Kalinowski, who joined OpenAI in November 2024 and led robotics and consumer hardware, announced her resignation on March 7, 2026. She said the company had rushed to announce the agreement before clear safeguards were in place. She specifically questioned surveillance of Americans without judicial oversight and lethal autonomous systems operating without human authorization. OpenAI reiterated that its red lines prohibit domestic surveillance and the use of AI to command autonomous weapons.