Key Points

  • Caitlin Kalinowski, OpenAI’s hardware leader, stepped down following the company’s agreement to deploy AI systems on Pentagon classified networks.
  • Her departure stemmed from worries about American citizen surveillance happening outside judicial oversight and autonomous lethal systems operating without human control.
  • She characterized the situation as a governance failure, noting the announcement preceded the establishment of protective measures.
  • The company maintains that its established boundaries prohibit domestic surveillance applications and autonomous weapon deployment.
  • This agreement followed Anthropic receiving supply chain risk designation due to their position on surveillance capabilities and lethal AI systems.

Caitlin Kalinowski, OpenAI’s hardware division leader, announced her resignation Saturday in response to the company’s recent partnership with the U.S. Department of Defense to deliver AI capabilities.

Kalinowski shared her decision through a post on X, explaining her departure as a matter of principle. Her tenure at OpenAI began in 2024, following her work directing augmented reality hardware initiatives at Meta Platforms.

In her statement, she expressed significant admiration for OpenAI CEO Sam Altman and her colleagues, while emphasizing her view that the Pentagon partnership moved forward too quickly.

“AI plays a critical role in national security,” she stated. “But surveillance of Americans without judicial oversight and lethal autonomy without human authorization are lines that deserved more deliberation than they got.”

She framed the matter as a governance challenge, pointing out that the partnership announcement occurred “without the guardrails defined.” She emphasized that decisions of such magnitude warrant careful consideration.

“These are too important for deals or announcements to be rushed,” she added in a subsequent X post.

The company countered her concerns by referencing its established boundaries that prohibit deploying its technology for domestic surveillance or autonomous weapons applications.

OpenAI provided a statement to Reuters indicating the company “will continue to engage in discussion with employees, government, civil society and communities around the world.”

Previous company statements emphasized that the Pentagon partnership incorporates enhanced safeguards governing AI model deployment.

The Path to OpenAI’s Pentagon Partnership

This agreement materialized after Anthropic, the AI firm supported by Google and Amazon investments, received supply chain risk designation. This classification connected to Anthropic’s established positions regarding surveillance capabilities and autonomous lethal systems.

Microsoft-backed OpenAI accepted the opportunity, committing to deploy its AI capabilities on the Pentagon’s classified cloud infrastructure.

Altman has emphasized the inclusion of safety measures in the agreement. The company’s Saturday statement reinforced this commitment.

Impact on OpenAI’s Executive Team

Kalinowski’s departure stems directly from her opposition to the partnership’s execution. She acknowledged the difficulty of her decision while expressing satisfaction with her robotics team accomplishments.

Reuters was unable to reach Kalinowski immediately for additional commentary.

Her resignation reveals friction within OpenAI regarding military partnerships. She represents one of the highest-ranking executives to depart over the Pentagon arrangement.

The company has yet to name a successor for the hardware oversight position.