A senior member of OpenAI’s robotics team resigned this week, citing concerns over the company’s recently announced partnership with the U.S. Department of Defense. Caitlin Kalinowski, a member of technical staff focused on robotics and hardware, announced her departure on social media, stating she stepped down “on principle” following the revelation that OpenAI would make its AI systems available within secure Defense Department computing environments.
Kalinowski’s resignation underscores a growing debate within the technology industry regarding the appropriate level of oversight and acceptable uses for advanced artificial intelligence in national security applications. In a series of public posts, she explained her decision, stating, “I resigned from OpenAI. I care deeply about the Robotics team and the work we built together. This wasn’t an simple call.”
She expressed concern that sufficient policy guardrails were not established before the agreement with the Pentagon was finalized. “AI has an important role in national security,” Kalinowski wrote. “But surveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got.”
Kalinowski emphasized her concerns centered on the process surrounding the decision, rather than specific individuals within OpenAI, stating she held “deep respect for Sam and the team, and I’m proud of what we built together,” referring to OpenAI chief executive Sam Altman.
An OpenAI spokesperson told NPR the company believes the 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 spokesperson added that OpenAI recognizes “people have strong views about these issues and we will continue to engage in discussion with employees, government, civil society and communities around the world.”
Kalinowski’s departure comes as competition intensifies among leading AI developers to secure contracts with the U.S. Government. In recent weeks, federal agencies have increasingly turned to OpenAI and Google for AI systems, following a dispute with Anthropic that led to the company being labeled a “Supply-Chain Risk to National Security” by Secretary of Defense Pete Hegseth, according to a report by Fox Business.
Anthropic’s CEO had publicly opposed allowing its software to be used for applications such as domestic mass surveillance or autonomous weapons, a position that clashed with defense officials who asserted the department requires flexibility to deploy commercial AI tools in all “lawful” operations. President Donald Trump subsequently ordered federal agencies to phase out their leverage of Anthropic technology, setting a six-month deadline.
Within OpenAI, Kalinowski focused on expanding the company’s robotics organization. Her LinkedIn profile indicates her work included hiring to support the company’s expansion into AI efforts related to physical infrastructure and machinery. She indicated plans to continue working in the field, stating, “I’m taking a little time, but I remain very focused on building responsible physical AI.”
The agreement between OpenAI and the Department of War (DoW) was announced late Friday, hours after a joint strike by the U.S. And Israel against Iran, according to a report by UPI. Altman publicly defended the deal on X, stating the DoW agrees with OpenAI’s principles prohibiting domestic mass surveillance and human responsibility for the use of force, including autonomous weapon systems.
In an all-hands meeting with OpenAI employees on Tuesday, Altman stated the Pentagon made clear to the company that it does not “get to make operational decisions” regarding how its AI technology is used, according to a partial transcript reviewed by CNBC. He used the examples of the Iran strike and a hypothetical invasion of Venezuela to illustrate that OpenAI employees do not have a role in determining military actions.
