Security Lunch: Wednesday, October 28th, 2026, 12:00PM @ CoDa E160
Title: Cryptographic Proof Systems for Scalable AI Oversight
Speaker: Ziyi Guan
Abstract: As AI systems become more capable, we need ways to make sure they do the right thing. This gives rise to a natural verification problem: can we check an AI system’s work much more efficiently than doing the work ourselves? Cryptography has
developed proof systems for exactly this kind of succinct verification. In this talk, I will discuss proof systems that have been proposed for scalable AI verification, including SNARGs and MIPs, and compare their advantages and limitations. Based on joint
works with Alessandro Chiesa, Yael Tauman Kalai, Zoe Xi, and Burcu Yıldız.
Bio: Ziyi Guan (https://ziyiguan.github.io/)
is a Swiss National Science Foundation (SNSF) Postdoc.Mobility Fellow and postdoctoral researcher at MIT and NYU. She is hosted by Yael Kalai and Vinod Vaikuntanathan at MIT, and Nir Bitansky and Benedikt Bünz at NYU. She received her Ph.D. in Computer Science
from EPFL in 2026, advised by Alessandro Chiesa and Mika Göös. Prior to that, she received bachelor's degrees in Computer Science and Mathematical Sciences from Carnegie Mellon University in 2020. Her research interests lie in theoretical computer science,
particularly probabilistic proof systems, cryptography, complexity theory, and quantum computing.