Alex Turner

Alex Turner is a research scientist at Google DeepMind on the scalable alignment team, where he works on training invariants into model behavior and related alignment problems. He previously formulated and proved the power-seeking theorems, co-formulated the shard theory of human value formation, and proposed the Attainable Utility Preservation (AUP) approach to penalizing negative side effects, and he now mentors researchers through the MATS program’s Team Shard stream.

No endorsements yet.

Personal Website

-

LinkedIn

-

Twitter / X

-

LessWrong

TurnTrout

EA Forum

Alexander Turner

No grants recorded.