Caitlin Kalinowski, former head of OpenAI’s robotics division, resigned after OpenAI signed a February agreement allowing its technology to be used across U.S. defense systems. She said concerns about domestic surveillance and lethal autonomy were not adequately addressed and called the announcement "rushed." OpenAI defended the pact, saying it enshrines red lines against domestic surveillance and autonomous weapons and pledges continued engagement with stakeholders. The resignation has sharpened debate over AI governance in wartime.
Top OpenAI Robotics Executive Resigns Over Controversial Pentagon Deal
A senior executive at OpenAI has resigned, citing deep concerns about a February agreement that allows the company’s technology to be used across U.S. defense and military systems.
Caitlin Kalinowski, who previously led OpenAI’s robotics division, announced her departure on social media, saying the deal raised serious governance questions about surveillance and lethal autonomy.
"This wasn’t an easy call," Kalinowski wrote. "AI has an important role in national security. But surveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got. This was about principle, not people."
Background
The Defense Department finalized an agreement with OpenAI in February while it was embroiled in a separate dispute with Anthropic, another AI startup. That conflict centered on restrictions set by Anthropic’s CEO, Dario Amodei, prohibiting the use of the company’s technology for mass domestic surveillance or for military attacks conducted without human oversight.
After talks stalled, President Donald Trump directed federal agencies in late February to stop using Anthropic’s technology. Last week the Pentagon designated Anthropic a supply-chain risk, a label typically reserved for organizations with ties to foreign adversaries.
Responses and Ongoing Debate
OpenAI confirmed Kalinowski’s departure and said its agreement with the Pentagon "creates a workable path for responsible national security uses of AI while making clear our red lines: no domestic surveillance and no autonomous weapons." The company added that it will continue engaging with employees, governments, civil-society groups and communities worldwide.
Kal inowski countered that the announcement was rushed and lacked clearly defined guardrails. Her resignation has intensified debate about how AI companies should govern dual-use technologies and who should set limits during wartime.
Reports indicate that AI tools have been used to support recent military operations, but there is no confirmed public instance of a U.S. strike executed without human oversight. Nonetheless, experts and advocates say the episode underscores the urgency of clear policies, transparency and judicial safeguards when AI is applied to national security.
Help us improve.
