A senior executive responsible for hardware development at OpenAI has stepped down after raising concerns about the company’s recent agreement with the United States Department of Defense. Caitlin Kalinowski, who led hardware initiatives at the artificial intelligence firm, announced her resignation on Saturday, stating that the company moved too quickly in approving the deployment of its artificial intelligence models on classified cloud networks used by the Pentagon. Her decision comes shortly after OpenAI confirmed its involvement in providing technology for defence related systems, a move that has sparked debate among employees and observers about the ethical and governance implications of such partnerships.
In a message posted on the social media platform X, Kalinowski said artificial intelligence can play an important role in national security but warned that certain uses of the technology require deeper discussion and oversight. She expressed concern that deploying advanced artificial intelligence systems within sensitive government infrastructure without extensive deliberation could raise serious governance issues. Kalinowski stated that potential risks such as surveillance activities without judicial oversight or the development of autonomous systems capable of making lethal decisions should have been considered more carefully before the agreement was announced. She added that the importance of these issues required stronger safeguards and a more transparent decision making process within the organisation.
Although Kalinowski noted that she continues to hold respect for Sam Altman, the Chief Executive Officer of OpenAI, and the broader team at the company, she indicated that the announcement of the Pentagon agreement came before sufficient guardrails had been clearly defined. In a subsequent statement, she described the matter primarily as a governance concern, emphasising that decisions involving powerful artificial intelligence technologies must be handled with greater institutional scrutiny. Reuters was not immediately able to reach her for further comment following her public statements.
OpenAI responded by reiterating that the agreement with the United States Department of Defense includes safeguards governing how its technology can be used. According to the company, its internal policies prohibit the use of its artificial intelligence systems for domestic surveillance activities or fully autonomous weapons systems. The company said it recognises that many stakeholders have strong views on the role of artificial intelligence in defence and national security and pledged to continue engaging with employees, government agencies, civil society groups and communities around the world on the responsible use of the technology.
Kalinowski joined OpenAI in 2024 after previously leading augmented reality hardware development at Meta Platforms, where she worked on advanced hardware initiatives related to immersive technologies. Her departure highlights growing debate across the technology sector about the relationship between private artificial intelligence developers and military institutions. As artificial intelligence capabilities expand, partnerships between technology firms and defence agencies are becoming more common, raising new questions about oversight, accountability and the ethical boundaries of advanced digital systems.
Follow the SPIN IDG WhatsApp Channel for updates across the Smart Pakistan Insights Network covering all of Pakistan’s technology ecosystem.
