Scott Emmons
Research Scientist at Google DeepMind
Scott received his PhD at UC Berkeley and was advised by Stuart Russell. He is interested in both the theory and practice of AI alignment. Scott has helped characterize how RLHF can lead to deception when the AI sees more than the human, develop multimodal attacks and benchmarks for open-ended agents, and use mechanistic interpretability to find evidence of learned look-ahead in a chess-playing neural network. Learn more at his website.