OpenAI’s robotics hardware lead, Caitlin Kalinowski, has resigned from the company following its recent deal with the Department of Defense, highlighting internal friction over AI safety and government partnerships.
Kalinowski’s departure marks a significant personnel change directly tied to OpenAI’s national security strategy. The resignation underscores the internal debates within the company regarding its approach to AI safety and collaborations with government entities. OpenAI’s agreement with the Pentagon is aimed at promoting responsible AI use in national security contexts.
Kalinowski expressed her concerns on X, stating that surveillance of Americans without judicial oversight and lethal autonomy without human authorization required more deliberation. She described the announcement of the DoD partnership as rushed and lacking defined guardrails, characterizing the issue as a “governance concern first and foremost.”
OpenAI responded to Engadget, acknowledging that people have strong views on these issues. The company clarified that it does not support the concerns raised by Kalinowski. OpenAI emphasized that its agreement with the Pentagon explicitly prohibits domestic surveillance and autonomous weapons.
Anthropic, another AI company, had previously refused to lift certain AI guardrails related to mass surveillance and fully autonomous weapons. OpenAI CEO Sam Altman stated that he would amend the DoD deal to prohibit spying on Americans. Kalinowski had previously worked at Meta before joining OpenAI in late 2024.
OpenAI confirmed Kalinowski’s resignation and stated that there are no plans to replace her. The company understands that people have strong views on these issues.
Correction, March 8, 2026: This story was updated to correct Kalinowski’s role to “robotics hardware lead” rather than “head of robotics.”
