Brian Jay Tang

Brian Jay Tang (LinkedIn) is a researcher and a Ph.D. candidate in Computer Science and Engineering at the University of Michigan, specializing in the security and privacy of AI systems and human-AI interaction. His research studies security for large language and vision-language models, privacy measurement at web scale, and user studies of how people understand and trust AI systems. His work has appeared at USENIX Security, IEEE Symposium on Security and Privacy, ACM CCS, PETS, ACM/IEEE HRI, and IMWUT/UbiComp, and has been covered by New Scientist and The Conversation. He has authored 13 papers, worked on grants totaling $1M+ in research funding from sponsors including NSF, DARPA, ARO, ONR, Samsung, and GM, and filed patents for systems related to agentic security and mobile screen protection. He presented his research on advertising in AI chatbots to the Federal Trade Commission, has guest lectured on AI security and privacy at Purdue CS and Michigan Law, and previously held roles at the IBM T.J. Watson Research Center, Roblox, and Optum. Brian conducted his undergraduate research at the University of Wisconsin–Madison, and is currently advised by Kang G. Shin at Michigan, where he was a College of Engineering Fellow.