Seeing us through machines: Designing AI to understand humans

Ziang Xiao, University of Illinois Urbana-Champaign
Host: Johns Hopkins Department of Computer Science

Many societal issues, such as health care, voting, etc., require decision-makers to study their stakeholders to design interventions or make a policy change. How do we conduct robust, generalizable, and engaging studies about human behavior? In this talk, I will share my vision on the role of AI in the quest of understanding humans and how could we approach such a future. I will introduce my work on designing and building conversational AI to conduct engaging surveys and collect high-quality information. I will first demonstrate the effectiveness of conversational AIs in transforming online survey experiences through a field study. Then, I will present a human-in-the-loop framework to create more effective interview chatbots with active listening skills. In the end, I will talk about my future research perspectives on designing and developing human-centered AI to understand humans for social change.

Speaker Biography

Ziang Xiao is a Ph.D. candidate in Computer Science at the University of Illinois Urbana-Champaign, advised by Prof. Hari Sundaram and Prof. Karrie Karahalios. He completed his B.S. in Psychology and Statistics & Computer Science at the University of Illinois Urbana-Champaign. His research lies at the intersection of human-computer interaction, natural language processing, and social psychology. The goal of his research is to enhance human-AI interactions to expand our understanding of human behavior. His work created engaging conversational agents to collect high-quality information through survey interviews. Ziang Xiao has published multiple papers in top-tier conferences and journals, including CHI, TOCHI, CSCW, IUI, etc.