Jonathn Chang

Hey! I’m a 3rd year Ph.D. student in the Center for Applied Mathematics at Cornell University, advised by Prof. Lionel Levine and Prof. Ziv Goldfeld. Previously, I earned my B.S. in Applied Mathematics from UCLA.

My research is broadly focused on AI alignment: ensuring that AI systems represent human values as they become more capable and autonomous. More specifically, I develop behavioral evaluations to measure the values expressed by large language models and use mechanistic interpretability to understand and steer the internal causal mechanisms that drive their behavior. Please feel free to reach out if you’re interested in chatting or collaborating!