A senior leader at OpenAI has stepped down following controversy surrounding the company’s agreement with the US Department of Defence. OpenAI Head of Robotics Caitlin Kalinowski announced her resignation after she raised objections about the company's recent AI contract with the Pentagon.
Kalinowski, who joined OpenAI from Meta in 2024, declared she made her decision based on her values. In a post on X she warned about potential risks related to AI surveillance and lethal autonomous systems.
She wrote that national security applications of artificial intelligence require further discussion because they involve two main issues: surveillance without judicial oversight and autonomous weapons that operate without human control.
The Pentagon agreement allows the US defence department to use OpenAI’s technology for certain applications. However, the partnership has triggered widespread debate about the boundaries of AI in military use.
However, in response to the criticism levelled against the agreement, an OpenAI spokesperson noted that the agreement has strict conditions. The company does not support surveillance in the home or autonomous weapons. It seeks to promote the use of AI in national security in a responsible manner.
This controversy also reveals a division between tech companies. Another AI firm, Anthropic, allegedly turned down a similar defence deal due to the possibility of the misuse of AI technologies.
Kalinowski’s departure could also impact OpenAI’s robotics research. OpenAI has been expanding its robotics programme for the past year with a dedicated robotics lab in San Francisco. The lab focuses on training robots to do household tasks as part of a long-term goal to develop a humanoid robot.