Paul Christiano, founder of the Alignment Research Center and former OpenAI safety researcher, has joined the OpenAI Foundation Board and its Safety and Security Committee.
This matters because Christiano is not a symbolic hire. He developed debate, amplification, and iterated amplification as concrete alignment techniques, and his work directly shaped how labs think about scalable oversight. Putting him inside OpenAI's governance structure is a structural change, not a PR move.
The details worth reading for: what specific authority the Safety and Security Committee holds, whether Christiano retains independence at ARC, and how this appointment interacts with OpenAI's ongoing restructuring into a public benefit corporation. The org chart question is the real story here.
[READ ORIGINAL →]