Principled Departure from OpenAI
Caitlin Kalinowski, who spearheaded OpenAI's advancements in robotics and consumer hardware, has announced her resignation from the prominent AI firm.
Her decision stems from profound reservations regarding OpenAI's recent agreement to deploy its AI models and tools on the Pentagon's secure, classified networks for military applications. Kalinowski expressed in a social media post that the swiftness with which OpenAI entered into this partnership, without sufficient deliberation, was the primary catalyst for her exit. She articulated that while AI holds undeniable significance for national security, the potential for unchecked surveillance of American citizens and the deployment of autonomous lethal weaponry without human authorization represent critical ethical lines that warranted more extensive discussion and consideration than they received. Kalinowski emphasized that her departure was driven by adherence to principles rather than any personal conflict with her colleagues, expressing deep respect for OpenAI CEO Sam Altman and the team's collective achievements.
AI Ethics in National Security
Kalinowski's resignation arrives at a critical juncture, underscoring a significant schism within the technology sector concerning the extent to which AI companies should support the US military's utilization of artificial intelligence. This situation mirrors a recent, high-profile dispute involving Anthropic, another leading AI developer. Anthropic's contract with the US Department of Defense was reportedly terminated after its CEO, Dario Amodei, refused to compromise on the company's established 'red lines.' These restrictions specifically prohibited their AI products from being employed for domestic surveillance within the United States or for powering fully autonomous armed drones. Shortly after Anthropic's contract dissolution, OpenAI finalized its own agreement with the Defense Department, a move that subsequently drew considerable criticism from both general users of ChatGPT and some OpenAI employees. Critics argued that this deal implied OpenAI had accepted terms that Anthropic had previously rejected, raising questions about ethical consistency and corporate responsibility in the rapidly evolving landscape of military AI.
OpenAI's Response and Clarifications
In the wake of the controversy, OpenAI CEO Sam Altman acknowledged that the company's announcement of the defense contract might have appeared 'opportunistic,' suggesting a recognition of the misstep in their communication strategy. Following this, OpenAI has actively worked to clarify the specific restrictions governing the military's use of its AI systems. A spokesperson for OpenAI conveyed to Fortune that their agreement with the Pentagon is designed to facilitate responsible national security applications of AI, while unequivocally maintaining their stated red lines: a strict prohibition against domestic surveillance and the development of autonomous weapons. The company has stated its awareness of the strong public opinions surrounding these issues and has committed to ongoing dialogue with employees, government entities, civil society organizations, and global communities to address these complex concerns and ensure responsible AI deployment in sensitive sectors.
Kalinowski's Background
Prior to her tenure at OpenAI, which began in November 2024, Caitlin Kalinowski amassed substantial experience in hardware development. She previously held a significant role at Meta, serving as a hardware executive for close to two and a half years. During her time with the social media giant, she was reportedly instrumental in the development of Meta's advanced Orion augmented reality glasses, previously known by the codename 'Project Nazare.' A prototype of these glasses was unveiled by Meta in September 2025. Furthermore, Kalinowski contributed to Meta's virtual reality initiatives for over nine years, including work on the Oculus headsets. Her earlier career also includes a nearly six-year stint at Apple, where she played a role in the design of MacBook models, including the Pro and Air lines. Her departure from OpenAI is viewed as a potential setback, particularly as the company is gearing up to launch its inaugural consumer hardware specifically designed for AI applications, a project developed in collaboration with renowned former Apple designer Jony Ive, with an anticipated release by the end of 2026.
