OpenAI’s head of robotics resigns over Pentagon deal, warning about surveillance and lethal autonomy

• OpenAI’s robotics chief, Caitlin Kalinowski, resigned after the company signed a defense contract with the Pentagon
• She said the agreement raised concerns about the surveillance of Americans and autonomous weapons
• The resignation highlights growing tensions inside the tech industry over AI’s role in military systems

OpenAI's head of robotics has stepped down after raising concerns about the company’s agreement with the U.S. Department of Defense. Caitlin Kalinowski announced that she had resigned following a deal between OpenAI and the Pentagon to deploy the company's AI models in certain government systems. At OpenAI, Kalinowski led efforts focused on robotics and physical systems, an area that many researchers believe will define the next major stage of AI.

But she felt compelled to leave regardless of that potential, driven by concerns about how quickly the agreement was reached and its possible implications for surveillance and autonomous weapons systems. She emphasized that her disagreement centered on governance rather than personal conflict.

“Surveillance of Americans without judicial oversight and lethal autonomy without human authorisation are lines that deserved more deliberation than they got,” she wrote.

Article continues below

Her resignation highlights the increasingly complicated relationship between cutting-edge AI companies and the national security establishment. The timing of OpenAI's deal made it particularly noteworthy, as it was announced only hours after rival AI company Anthropic reportedly refused to authorize broad military uses of its own models.

Anthropic’s decision triggered a strong response from government officials, who subsequently designated the company as a supply chain risk after it declined to provide unrestricted access to its technology.

Deals involving national security infrastructure typically involve lengthy negotiations and careful oversight. The rapid turnaround raised questions about OpenAI's own diligence. Kalinowski’s public comments echoed those concerns. In a follow-up message explaining her resignation, she said the problem was not the concept of a defense partnership itself, but the pace at which the decision to move forward was made.

Kalinowski's role with robotics made the Pentagon deal especially significant, as autonomous systems and robots all have potential military applications. Her departure nevertheless underscored the tension that sometimes arises when advanced technology meets national security priorities.

Rushed AI deals

That said, OpenAI CEO Sam Altman has tried to calm the waters. He stated that the contract would be adjusted to ensure the company’s models could not be used for domestic surveillance of U.S. citizens. While OpenAI has gestured towards opposing the development of fully autonomous lethal systems, having to say so doesn't exactly inspire confidence.

Some government agencies increasingly view AI as a strategic capability that could shape the future balance of power. But some tech firms are uneasy about how closely they should collaborate with the military. Or if they aren't, they're at least uneasy about how their regular customers would react to them working with the military.

Kalinowski’s resignation is unlikely to derail OpenAI’s defense partnership, but it might at least prompt further questions and perhaps slow things down a little. Some companies will conclude that collaboration with government agencies is necessary to ensure democratic oversight and the responsible use of advanced technology. But it may also remind the industry that decisions about how AI is used in national security deserve careful scrutiny.

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button!

And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.