OpenAI robotics head Caitlin Kalinowski resigned following the company’s AI deal with the US Department of War.
She said the decision was based on principles and concerns about how AI could be used, especially in surveillance or weapons.
The deal has also sparked backlash online, with some users protesting and switching to rival AI apps.
OpenAI has been losing significant public support following its recent deal with the Pentagon. Initially, the company faced strong criticism from users over what many described as a rushed agreement, which also triggered the QuitGPT backlash online. Now, Caitlin Kalinowski, Head of Robotics and Consumer Hardware at OpenAI, has resigned. She shared her decision on the social media platform X (formerly Twitter), where she said that the decision was hard but necessary. Kalinowski explained that her decision was connected to OpenAI’s recent deal with the United States Department of War (DoW). She said artificial intelligence can help with national security but warned that using AI for surveillance and weapons needs strict rules and careful oversight. Her resignation comes just days after OpenAI confirmed that its AI models could be deployed within classified United States Department of War systems.
Kalinowski joined OpenAI in late 2024 after leading augmented-reality hardware projects at Meta. Earlier in her career she worked at Apple, contributing to several MacBook designs. Her exit raises fresh questions about the future direction of OpenAI’s robotics efforts and the broader debate over how artificial intelligence should be used in military settings.
Also read: Samsung Galaxy S27 Ultra specs tipped days after Galaxy S26 Ultra launch: What to expect
In her post, Kalinowski said her decision to leave was based on principle rather than disagreements with individuals at the company. She added that she has deep respect for OpenAI chief executive Sam Altman and the team she worked with and said she remains proud of the robotics work completed during her tenure.
However, she argued that the Pentagon agreement appeared rushed and was announced without clearly defined safeguards. Kalinowski said issues such as potential surveillance of Americans without judicial oversight and the risk of lethal autonomous systems warrant broader public debate before such partnerships are finalised. She later clarified that her concerns focused primarily on governance and the absence of clear guardrails at the time of the announcement.
Also read: Amazon Electronics Premier League 2026: Apple iPhone Air available with over Rs 29,400 discount
OpenAI confirmed her departure and defended the defence agreement, saying the arrangement provides a ‘responsible pathway’ for national security applications of AI while maintaining strict limitations. The company said its technology will not be used for domestic surveillance of US citizens or fully autonomous weapons.
The deal follows earlier discussions between the Pentagon and rival AI firm Anthropic. Those talks reportedly broke down after Anthropic pushed for stronger restrictions on mass surveillance and autonomous weapons. The War Department subsequently labelled Anthropic a supply chain risk, a designation the company has said it intends to challenge in court. Anthropic chief executive Dario Amodei has also questioned the legal basis of the move.
The controversy has also affected public sentiment as some users uninstalled ChatGPT in protest while downloads of Anthropic’s chatbot Claude surged, briefly pushing it to the top of Apple’s US App Store rankings.
