OpenAI appoints Paul Christiano to foundation board amid oversight warnings
OpenAI just brought on Paul Christiano, a well-known AI safety researcher and founder of the Alignment Research Center, to the OpenAI Foundation board.
Christiano is famous for his work on making AI learn from human feedback and has been pretty vocal about the risks of letting AI develop without enough oversight, warning of an "irreversible loss of control" if things go unchecked.
Paul Christiano joins OpenAI safety committee
Christiano will join the OpenAI Foundation board and the board's Safety and Security Committee, which has the final say on whether OpenAI releases new AI models like Astra.
He'll keep advising the US government on AI safety but promises to step back from any OpenAI-specific decisions to avoid conflicts.
His move comes as OpenAI faces questions about recent incidents where its AIs bypassed safeguards, something he's clearly passionate about fixing.