OpenAI has appointed Paul Christiano to its Foundation Board, a move that signals a deepening commitment to rigorous safety governance as AI capabilities accelerate. Christiano will serve as a non-voting observer on the OpenAI Group PBC Board and join the Safety and Security Committee (SSC), working directly with committee chair Zico Kolter. This appointment places one of the field’s most respected alignment researchers at the heart of OpenAI’s oversight structure, bridging the gap between technical research and corporate governance.
Christiano is no stranger to the organization. He led alignment research at OpenAI from 2017 to 2021, contributing foundational work on reinforcement learning from human feedback (RLHF), the technique that underpins much of modern chatbot behavior. Since leaving, he founded the Alignment Research Center (ARC) and served as a Senior Tech Advisor at the U.S. AI Safety Institute within NIST. His recent government role involved evaluating frontier models for national security risks, giving him a rare dual perspective on both technical alignment and policy-level safety standards.
The OpenAI Foundation explicitly notes that Christiano brings a perspective that “takes seriously the possibility that advanced AI could pose catastrophic risks.” This is not a ceremonial hire. The Foundation states it seeks individuals prepared to “challenge prevailing assumptions” and strengthen the evidence behind critical safety decisions. By placing an independent voice known for skepticism about industry safeguards into a formal governance role, OpenAI appears to be responding to growing pressure for more robust accountability mechanisms following its recapitalization in October 2025.
This structural shift builds on governance reforms mandated by reviews from the California and Delaware Attorneys General. The Foundation, which holds significant equity in and control over OpenAI Group PBC, uses committees like the SSC to oversee safety practices across the entire organization. Christiano’s addition suggests that OpenAI views technical alignment not just as a research problem, but as a board-level fiduciary duty. Whether this internal check can keep pace with external regulatory demands remains the open question.
