Jeremy is a professor in the Department of Philosophy and the Department of Mathematical Sciences at Carnegie Mellon University. His research focuses on mathematical logic and formal methods in mathematics, as well as the philosophy and history of mathematics.

Samy Bengio is a Senior Directory of AI and Machine Learning Research at Apple. His recent work has studied the reasoning abilities of transformers, particularly from a mathematical and theoretical perspective. Before that, he was a distinguished scientist at Google Research since 2007 where he was heading the Google Brain team, and at IDIAP in the early 2000s where he co-wrote the well-known open-source Torch machine learning library.

Noam Brown is a research scientist at OpenAI working on multi-step reasoning, self-play, and multi-agent AI. He previously developed CICERO, the first AI to achieve human-level performance at Diplomacy and previously created the first AI to defeat top humans in no-limit poker.

Junehyuk Jung is an associate professor of mathematics at Brown University, and a Visiting Researcher at Google DeepMind. He is interested in problems involving arithmetic hyperbolic manifolds and/or eigenfunctions of the Laplace-Beltrami operator.

Dawn Song is a Professor at UC Berkeley. She has done research in AI and deep learning, blockchain/web3, security and privacy, as well as AI for coding and mathematical reasoning. She is the recipient of various awards including the MacArthur Fellowship, the Guggenheim Fellowship, and the NSF CAREER Award.

Adam Wagner is an assistant professor of mathematics at WPI and a mathematical consultant for Google DeepMind. His former focus was in combinatorics and graph theory, though his recent work has shifted towards finding ways to integrate AI with mathematics research to aid mathematical intuition and advance mathematical research.

Denny is the founder and lead of the Reasoning Team in Google Brain, aiming to revolutionize machine learning by introducing reasoning to address challenges such as learning from few examples or instructions only, enhancing interpretability, and handling out-of-distribution/domain generalization.