Scott Emmons previously cofounded FAR.AI and served as a Research Advisor. He is now a research scientist at Google DeepMind focused on AI safety and alignment. He is wrapping up his PhD at UC Berkeley’s Center for Human-Compatible AI, advised by Stuart Russell. He is interested in both the theory and practice of AI alignment. He has helped characterize how RLHF can lead to deception when the AI sees more than the human, developed multimodal attacks and benchmarks for open-ended agents, and used mechanistic interpretability to find evidence of learned look-ahead in a chess-playing neural network. For more information, visit his website.