OpenAI has named Paul Christiano, a prominent AI safety researcher who previously led alignment work at the company before leaving in 2021, to its foundation board. He will also sit on the board’s Safety and Security Committee.
Christiano spent his early career focused on keeping advanced models aligned with human intent, first as a researcher at OpenAI and later as founder of the Alignment Research Center, where he studied how to verify and control systems far smarter than their builders. His return to a governance seat reunites the lab with one of the few outside voices it has historically struggled to keep close.
The appointment gives OpenAI’s nonprofit arm a director with deep technical credibility on exactly the questions regulators and rivals keep pressing: what happens when models can improve themselves, and who decides when a training run is too risky to start. OpenAI said in its announcement that Christiano brings experience in alignment, safety and standards work to the board.
The move follows a period of intense scrutiny of OpenAI’s safety governance, including agent incidents over the summer and public disagreements inside the lab about the pace of development. Adding an independent-minded alignment specialist to the oversight layer reads as an attempt to signal that safety voices have real influence over company decisions.
