Shi Feng

• Program

  Science of Trustworthy AI

• Institution

  The George Washington University

• Location

  USA

Shi leads a research group working on technical AI safety at George Washington University where he is an assistant professor. He recently focuses on the emergence of deception and collusion, investigating meta-cognitive capabilities as a driving force behind them. He designs benchmarks and algorithmic interventions to evaluate and mitigate safety risks. Before GWU, Shi did postdocs at UChicago and NYU, and received his PhD from UMD.