OpenAI robotics chief resigns over Pentagon contract, citing surveillance and autonomous weapons concerns
Key Points
- Caitlin Kalinowski, OpenAI’s robotics chief, resigned after a Pentagon contract was announced.
- She cited concerns about surveillance of Americans without judicial oversight.
- She warned that the deal could enable lethal autonomous weapons lacking human authorization.
- Kalnowski emphasized that the speed of the agreement left insufficient time for governance review.
- OpenAI CEO Sam Altman said the contract would be modified to block domestic surveillance use.
- The resignation highlights growing tension between AI firms and national‑security priorities.
- A rival AI firm, Anthropic, recently declined broad military use of its models, drawing government criticism.
- Kalnowski’s departure may prompt other tech companies to scrutinize the pace and transparency of defense deals.
OpenAI's head of robotics, Caitlin Kalinowski, stepped down after the company signed a defense agreement with the U.S. Department of Defense. Kalinowski said the rapid deal raised serious governance issues, particularly the potential for domestic surveillance without judicial oversight and the use of AI in lethal autonomous systems. While OpenAI CEO Sam Altman assured that safeguards would be added, the resignation highlights growing tension between cutting‑edge AI firms and national‑security priorities.
Resignation Announcement
Caitlin Kalinowski, who led OpenAI’s robotics and physical‑systems efforts, announced her resignation following the company’s newly announced contract with the Pentagon. The agreement would place OpenAI’s AI models into certain government systems, a move that Kalinowski described as being made too quickly and without sufficient deliberation.
Concerns Over Surveillance and Lethal Autonomy
Kalnowski’s departure centered on two primary concerns. First, she warned that the deal could enable surveillance of Americans without judicial oversight. Second, she highlighted the risk that the technology could be used in lethal autonomous weapons without human authorization. She framed these as governance issues rather than personal disagreements, emphasizing that the speed of the announcement left insufficient time for careful policy review.
Reaction from OpenAI Leadership
OpenAI chief executive Sam Altman responded by stating the contract would be adjusted to prevent the models from being used for domestic surveillance of U.S. citizens. He also reiterated OpenAI’s opposition to fully autonomous lethal systems, though the public statement did not fully assuage concerns about the partnership’s oversight.
Industry Context and Implications
The resignation comes amid heightened scrutiny of AI firms’ relationships with the national‑security establishment. Earlier that day, a rival AI company, Anthropic, reportedly declined to provide unrestricted access to its models for broad military use, prompting a government response that labeled the company a supply‑chain risk. Kalinowski’s exit underscores the tension that can arise when advanced AI capabilities intersect with defense priorities and may prompt other firms to reconsider the pace and transparency of similar agreements.
Future Outlook
While Kalinowski’s resignation is unlikely to halt OpenAI’s defense partnership, it may slow the rollout and encourage more robust internal governance. The episode serves as a reminder that decisions about AI deployment in national‑security contexts demand careful oversight and public accountability.