‹ Back

OpenAI Hardware Chief Resigns After Pentagon AI Deal

 

OpenAI hardware leader Caitlin Kalinowski resigned after the company agreed to provide AI services to the U.S. Department of Defense.

 

 

OpenAI Hardware Leader Steps Down Following Pentagon Agreement

 

OpenAI’s head of robotics and consumer hardware, Caitlin Kalinowski, has resigned following the company’s agreement to provide artificial intelligence technology to the U.S. Department of Defense, according to statements from the executive and confirmation from the company.

 

Kalinowski announced her departure in a social media post, stating that she was stepping down over concerns related to the company’s recently announced deal with the Pentagon. She wrote that while artificial intelligence has a role in national security, the agreement raised issues that required deeper consideration before implementation.

 

“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.”

 

The resignation comes shortly after OpenAI confirmed that its AI models would be deployed within classified U.S. Defense Department computing environments, part of a broader collaboration aimed at supporting national security applications.

 

Concerns Over Governance and Safeguards

 

In additional remarks posted on social media, Kalinowski said the agreement had been announced “without the guardrails defined,” describing the issue as primarily a governance concern. She added that decisions involving sensitive national security applications of artificial intelligence required careful oversight and should not be rushed.

 

Kalinowski also emphasized that her decision was not directed at colleagues within the organization, stating that she maintained “deep respect” for OpenAI Chief Executive Sam Altman and the company’s employees.

 

OpenAI confirmed her departure in statements to media outlets and reiterated that the agreement with the Pentagon contains restrictions on how its technology may be used. According to the company, the partnership establishes explicit “red lines” prohibiting the use of OpenAI systems for domestic surveillance or autonomous weapons.

 

The company added that it intends to continue discussions with employees, government institutions, civil society groups, and other stakeholders regarding the responsible use of artificial intelligence in national security settings.

 

Pentagon Collaboration and Defense Applications

 

OpenAI has said its technology could support several defense-related functions, including cybersecurity, intelligence analysis and logistics operations. Such applications have been identified by military planners as areas where artificial intelligence could improve operational efficiency and data analysis capabilities.

 

The agreement allowing OpenAI’s systems to operate in classified environments was announced shortly after discussions between the Pentagon and the artificial intelligence company Anthropic failed to produce a contract.

 

The shift opened the way for OpenAI to expand its cooperation with the Department of Defense. The company had previously been among several technology firms—including Anthropic, Google and xAI—that participated in a $200 million Pentagon initiative announced in June to develop AI tools for national security purposes.

 

Kalinowski’s Role at OpenAI

 

Kalinowski joined OpenAI in 2024 after previously leading augmented-reality hardware development at Meta Platforms. At OpenAI she oversaw hardware initiatives and robotics development, including work related to the company’s consumer hardware and robotics programs.

 

In her resignation statement, Kalinowski said the decision was difficult but described it as a matter of principle tied to how advanced artificial intelligence technologies should be deployed in security contexts.

 

Her departure marks one of the most senior exits connected to the company’s recent defense partnership announcement, which has drawn scrutiny from some employees and AI researchers over how rapidly the agreement was announced and the safeguards governing its implementation.

 

OpenAI has maintained that the Pentagon partnership is structured to enable national security uses of artificial intelligence while maintaining restrictions designed to prevent surveillance abuses and the development of fully autonomous weapons systems.