Roger Grosse 2024 Senior Fellow
Affiliation Associate Professor, University of Toronto Hard Problem Solved AI’s continually evolving safety and security, robustness, performance, output challenges and other shortcomings that may cause harm or erode public trust of AI systems, especially in safety-critical applications and uses where societal stakes and potential for societal harm are high.

Roger Grosse is an Associate Professor of Computer Science at the University of Toronto, Schwartz-Reisman Chair in Technology and Society, founding member of the Vector Institute for Artificial Intelligence, and a Member of Technical Staff on the Alignment Science team at Anthropic. His research focuses on using our understanding of deep learning to improve the safety and alignment of AI systems. He has held the Sloan Research Fellowship, Canada CIFAR AI Chair, and Canada Research Chair.

AI2050 Project

With the rapid progress in AGI capabilities, AGI systems will learn potentially catastrophic capabilities, and developers will need to be able to demonstrate their safety. Unfortunately, we lack the algorithmic techniques that would be needed to make safety cases. Roger’s AI2050 project will address two algorithmic problems needed for safety cases: first, to determine what properties of a model have changed as the result of a stage of training; and second, to either find or bound the probability of rare behaviors for a model (such as executing malign plans).