Shi Feng
Principal Investigator
Shi Feng is an Assistant Professor at George Washington University where his research studies AI safety and alignment. Focusing on future AI systems that are potentially more capable than humans, his main goal is to empower human oversight and inform policy decisions.
He designs new theories, algorithms, and user interfaces to augment human decisions with and around AI systems. His work on scalable oversight extends from his research on human-AI collaboration, interpretability evaluation, and adversarial robustness.
Shi received his PhD at UMD supervised by Jordan Boyd-Graber. He did postdocs at UChicago with Chenhao Tan and NYU in the Alignment Research Group with Sam Bowman and He He.
His most recent work focuses on meta-evaluation of risks in scalable oversight methods. As AIs are deployed to solve increasingly complex problems, reliable human oversight becomes a huge challenge: AIs are getting better at producing outputs that look correct to humans but are in fact subtly flawed. Shi leads a research group working on oversight and control to address these challenges.
He has published extensively with over 5,800 citations and serves as Assistant Professor in GWU's Co-Design of Trustworthy AI Systems program.
Begin Your Journey
The application takes 10 minutes and is reviewed on a rolling basis. We look for strong technical signal—projects, coursework, or competition results—and a genuine curiosity to do real research.
If admitted, you will join a structured pipeline with direct mentorship to take your work from ideation to top conference submission at venues like NeurIPS, ACL, and EMNLP.
