Full Bio - CV - Publications

Bio

Kellin Pelrine is a Member of Technical Staff at FAR.AI. He founded and leads the technical parts of FAR.AI’s red team. Previously he cofounded Stitch, Mutual, and Machine Learning Sports, and completed a PhD at McGill University and the Mila AI Institute (supervised by Reihaneh Rabbany and Jean-François Godbout). He also holds a master’s in economics from Yale University and a master’s in applied mathematics from the University of Colorado at Boulder.

Pelrine leads cross-functional teams towards secure, transformatively impactful AI systems. His current work spans uncovering, understanding, and solving AI risks across misuse, misalignment, and loss of control. In the past he has worked across a wide range of problems in AI safety and security, education, persuasion and manipulation, and temporal graphs. His work has been published and won awards in top venues such as NeurIPS, ICML, AAAI, KDD, and TMLR, and featured in major media outlets including The Financial Times, The Times, Vice, Scientific American, and Nature. It has also been cited in the US Senate, highlighted in Canada’s National AI Strategy, and used by governments globally. His findings and recommendations have been deployed and improved safeguards in models from every frontier AI company.

When time allows, he is a foodie, and an avid player and teacher of the board game Go. He has twice represented the US in international competition, including the World Mind Sports Games. For beating “superhuman” Go AIs as a human, he has been called the man who beat the machine.

back