September 10, 2026, (Inside AI) — Paul Christiano, a former OpenAI alignment lead and founder of the Alignment Research Center, has joined the OpenAI Foundation Board as a non-voting observer on the OpenAI Group PBC Board. He will also serve on the Foundation’s Safety and Security Committee, working alongside chair Zico Kolter.
The appointment places one of the field’s most prominent voices on catastrophic AI risk inside OpenAI’s governance structure. Christiano’s career spans government service and independent research, giving him a rare vantage point on how frontier labs handle safety.
Christiano currently works as a Senior Tech Advisor at the Center for AI Standards and Innovation within NIST, part of the U.S. Department of Commerce. His government experience covers two administrations, including work at CAISI’s predecessor, the U.S. AI Safety Institute. There he evaluated frontier AI models for national security implications and mitigation strategies.
Before government service, Christiano led alignment research at OpenAI from 2017 to 2021. He contributed foundational work on reinforcement learning from human feedback, a technique now central to training large language models. His nonprofit, the Alignment Research Center, focuses on aligning advanced AI systems with human interests.
OpenAI’s announcement frames the appointment as a governance strengthening move. The Safety and Security Committee provides oversight across all of OpenAI, including OpenAI Group PBC. The Foundation controls the PBC and holds a significant equity stake in it.
The Foundation’s statement said: “We believe the Foundation’s governance is stronger with people prepared to challenge prevailing assumptions and help strengthen the evidence, safeguards, and accountability underlying critical decisions.”
Christiano’s appointment follows OpenAI’s recapitalization in October 2025. That restructuring created the Foundation as a charitable organization controlling the for-profit PBC. Reviews by the California Attorney General and Delaware Attorney General influenced the governance framework.
Why Christiano’s Appointment Signals a Governance Shift
Christiano has publicly questioned whether industry safeguards are adequate for advanced AI. His research takes seriously the possibility that AI systems could pose catastrophic risks before reaching dangerous capability levels. That stance contrasts with some industry narratives emphasizing rapid deployment.
His dual role as government advisor and independent researcher adds weight to the Foundation’s oversight function. The Safety and Security Committee now includes someone who has evaluated frontier models for national security risks from inside NIST.
The non-voting observer status on the PBC board limits his direct control over commercial decisions. But committee membership gives him influence over safety governance across OpenAI’s operations. This structure mirrors governance models used in other high-risk industries where independent oversight matters.
OpenAI’s recapitalization drew scrutiny from state attorneys general over charitable asset transfers and governance safeguards. Christiano’s appointment appears designed to address concerns about independent safety oversight. The Foundation’s charitable mission includes advancing scientific discovery and strengthening civil society through responsible AI.
What Christiano Brings to the Table
Christiano’s RLHF work helped shape modern AI training methods. His later research at ARC focused on evaluating whether AI systems could deceive human evaluators or resist shutdown attempts. That technical grounding distinguishes him from board members with primarily business or policy backgrounds.
At NIST, Christiano worked on evaluating frontier AI models with national security implications. That experience includes assessing capabilities that could be misused by adversaries. His appointment creates a direct channel between federal AI safety evaluation work and OpenAI’s internal governance.
The Foundation’s announcement emphasizes Christiano’s willingness to challenge assumptions. His public statements have warned that developers must prepare safety measures before systems reach dangerous capability levels. That perspective aligns with the Foundation’s stated mission to ensure AGI benefits all of humanity.
Christiano’s appointment does not change OpenAI Group PBC’s day-to-day operations directly. But it strengthens the Foundation’s ability to exercise oversight through the Safety and Security Committee. The committee’s governance role covers safety and security practices across all OpenAI entities.
The Foundation supports grants, partnerships, and philanthropic initiatives to expand access to AI’s benefits. Christiano’s technical expertise could inform how those initiatives evaluate safety implications. His government connections may also facilitate coordination with federal AI safety efforts.
OpenAI has not disclosed the term length or specific voting rights for Christiano’s observer role. The appointment adds another layer to a governance structure still evolving after the 2025 recapitalization. Industry observers will watch whether Christiano’s independent voice translates into concrete safety policy changes.
