Story
September 10, 2026
OpenAI Brings Alignment Researcher Paul Christiano Back Into the Boardroom
OpenAI is presenting Paul Christiano’s appointment as a practical strengthening of safety governance, while the label attached to him elsewhere underscores how central—and contested—warnings about advanced AI have become.
OpenAI has appointed Paul Christiano, a former employee and one of the field’s best-known alignment researchers, to the boards of both the company and its nonprofit foundation. The decision places a specialist in the problem of keeping powerful AI systems aligned with human interests closer to the organization’s formal oversight machinery.1
Christiano worked at OpenAI from 2017 to 2021, leading alignment research and contributing to reinforcement learning from human feedback, or RLHF—the technique used to steer models using human preferences. After leaving, he founded the Alignment Research Center and served as a senior technical adviser to the federal Center for AI Standards and Innovation.1
The new role is not merely ceremonial. Christiano will sit on the OpenAI Foundation board’s Safety and Security Committee alongside chair Zico Kolter, a committee that governs safety practices at the for-profit company. Bret Taylor, OpenAI’s board chair, said Christiano had helped define AI alignment through work focused on “the hardest questions posed by increasingly capable systems.”1
Christiano framed the appointment as an answer to a widening technical challenge. “AI capabilities have advanced very rapidly in the last year and alignment remains a difficult technical problem,” he said, adding that the committee’s job is becoming “more important and more challenging than ever.”1
OpenAI’s leadership publicly welcomed the return. Co-founder Greg Brockman wrote, “welcome @paulfchristiano!” while amplifying the foundation’s safety-governance role.
2 Chief executive Sam Altman called Christiano’s safety work valuable and said he was “Excited to work together again.”
3
The appointment is likely to be read through a sharper outside lens, too: another account described Christiano as a “prominent AI doomer,” reflecting the more alarmed camp’s concern that capability gains are racing ahead of reliable safeguards.4 OpenAI, by contrast, is signaling that those concerns belong inside its governing structure—not at its margins.