SAN FRANCISCO: The head of OpenAI’s robotics division has resigned, citing concerns over the company’s agreement with the United States Department of Defence to deploy its artificial intelligence models within the Pentagon’s classified networks.
Caitlin Kalinowski, who oversaw hardware and robotics development at the AI startup, announced her resignation on Saturday in a post on social media platform X.
“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 authorisation are lines that deserved more deliberation than they got.”
I resigned from OpenAI. I care deeply about the Robotics team and the work we built together. This wasn’t an easy call. AI has an important role in national security. But surveillance of Americans without judicial oversight and lethal autonomy without human authorization are…
— Caitlin Kalinowski (@kalinowski007) March 7, 2026
Concerns over AI governance
Kalinowski said her decision was rooted in concerns about governance and the pace at which the agreement with the Pentagon was announced.
“It’s a governance concern first and foremost,” she wrote in a subsequent post. “These are too important for deals or announcements to be rushed.”
She said OpenAI had not taken sufficient time to define safeguards before agreeing to deploy its AI models on the Pentagon’s classified cloud networks.
Despite her decision to step down, Kalinowski said she had “deep respect” for OpenAI Chief Executive Sam Altman and the company’s team.
“This was about principle, not people,” she wrote.
Kalinowski joined OpenAI in 2024 after leading augmented reality hardware development at Meta Platforms.
Pentagon agreement
OpenAI confirmed Kalinowski’s departure but defended the agreement with the Pentagon.
In a statement to Reuters, the company said the deal provides a path for responsible use of AI in national security while maintaining clear restrictions.
“Our 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,” OpenAI said.
The company added that it recognises strong views on the issue.
“We recognise that 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,” the statement said.
OpenAI also said the agreement includes additional safeguards designed to prevent misuse of its technology.
Debate over military AI
The deal comes amid growing debate over the role of artificial intelligence in military operations and surveillance.
According to reports, the Pentagon agreement allows OpenAI’s technology to be deployed across parts of the US defence and national security infrastructure.
The agreement was reached as the US Defence Department was engaged in a dispute with another AI startup, Anthropic, over restrictions placed on the use of its technology.
Anthropic’s chief executive, Dario Amodei, had barred the use of its AI models in mass surveillance of US citizens or in military attacks without human input.
US President Donald Trump directed federal agencies in late February to stop using Anthropic technology.
The Pentagon subsequently designated the company a supply chain risk, a label typically reserved for entities linked to foreign adversaries.
