OpenAI introduced on September 9, 2026 the appointment of Paul Christiano to the OpenAI Basis Board, the place he’ll be a part of the board’s Security and Safety Committee. Christiano may also function a non-voting observer on the board of OpenAI Group PBC.
Security and Safety Committee Function
Christiano will work on the Security and Safety Committee alongside its chair, Zico Kolter. The committee gives governance over security and safety practices throughout all of OpenAI, together with OpenAI Group PBC. A footnote to the announcement states that, as a Senior Technical Advisor, Christiano will recuse himself from all OpenAI-related issues in addition to all mannequin evaluations.
Bret Taylor, chair of the OpenAI Basis and OpenAI Group PBC boards, mentioned Christiano’s expertise engaged on frontier AI security and requirements in authorities, collectively together with his technical judgment, will strengthen the board’s oversight and the committee’s work as OpenAI continues to develop and deploy frontier AI. “Paul has helped outline the sphere of AI alignment by work that’s rigorous and centered on the toughest questions posed by more and more succesful methods,” Taylor mentioned.
Authorities and Alignment Analysis Background
Christiano is a Senior Tech Advisor on the Heart for AI Requirements and Innovation inside the Nationwide Institute of Requirements and Expertise, an company of the U.S. Division of Commerce, with authorities expertise spanning two administrations. Throughout his time on the heart and its predecessor, the U.S. AI Security Institute, he has labored on evaluating frontier AI fashions, together with capabilities with nationwide safety implications, and on approaches to mitigating the related security and safety dangers.
Christiano can be the founding father of the Alignment Analysis Heart, a nonprofit analysis group centered on aligning superior AI methods with human pursuits. From 2017 to 2021, he led alignment analysis at OpenAI and contributed foundational work on reinforcement studying from human suggestions.
In keeping with OpenAI, Christiano’s profession researching AI alignment displays a perspective that treats the opportunity of catastrophic dangers from superior AI severely and holds that builders should put together earlier than methods attain extra harmful ranges of functionality. The corporate described him as an unbiased voice on whether or not the trade’s safeguards are enough, and mentioned it believes the Basis’s governance is stronger with folks ready to problem prevailing assumptions and assist strengthen the proof, safeguards, and accountability underlying crucial choices.
“AI capabilities have superior very quickly within the final 12 months and alignment stays a tough technical drawback, making the Security and Safety Committee’s duty extra vital and more difficult than ever,” Christiano mentioned, including that he’s excited to hitch the group to assist.
Governance Construction From the 2025 Recapitalization
OpenAI mentioned the appointment builds on the governance construction established by its recapitalization in October 2025, together with commitments made in reference to evaluations performed over the course of a 12 months by the California Lawyer Common and the Delaware Lawyer Common, whose engagement the corporate mentioned helped strengthen the governance framework.
Underneath that construction, introduced on October 28, 2025 and detailed on OpenAI’s construction web page, the nonprofit turned the OpenAI Basis and the for-profit turned OpenAI Group PBC, a public profit company required to advance its said mission and take into account the broader pursuits of all stakeholders. OpenAI was based in 2015 as a nonprofit with a mission to make sure that synthetic normal intelligence advantages all of humanity, and it created the for-profit subsidiary in 2019. Via particular voting and governance rights held solely by the Basis, the Basis appoints all members of the OpenAI Group board and may substitute administrators at any time.
The construction web page lists the Basis board as unbiased administrators Bret Taylor as chair, Adam D’Angelo, Christiano, Sue Desmond-Hellmann, Zico Kolter, retired U.S. Military Common Paul M. Nakasone, Adebayo Ogunlesi, Nicole Seligman, David Vélez, and Robin Vince, together with CEO Sam Altman. Kolter, the Security and Safety Committee’s chair, serves completely on the Basis board and as a non-voting observer to the Group board, whereas the opposite present Basis administrators additionally serve on OpenAI Group’s board. Inside one 12 months of the recapitalization, a second Basis director will transition to serving completely on the Basis board and as a non-voting observer to the Group board.
Basis Operations and Fairness
The announcement describes the OpenAI Basis as a charitable group that operates as a separate nonprofit, conducting its personal charitable operations whereas sustaining management of OpenAI Group PBC and holding a big fairness stake in it. The Basis helps initiatives advancing scientific discovery, strengthening civil society, and constructing resilience by accountable AI, and its web site lists three preliminary precedence applications: Life Sciences and Curing Ailments, AI Resilience, and Civil Society and Philanthropy.
As of the closing of the recapitalization, the Basis holds a 26% fairness stake in OpenAI Group, value roughly $130 billion primarily based on OpenAI Group’s present valuation, in accordance with the construction web page. The Basis additionally holds a warrant permitting it to obtain extra shares if OpenAI Group reaches a valuation milestone; if OpenAI Group’s share worth will increase by greater than tenfold after 15 years, the Basis will obtain important extra fairness. Microsoft holds roughly 27% of OpenAI Group as of the closing, with the remaining 47% held by present and former staff and buyers.
