Topic: Strategic Interactions with the Assistant of an Algorithm: the Power of Data and Mechanism
Speaker: Lijia Wei
Time: October 26, 2022 15:00
Location: Tencent meeting, ID: 953-473-375
Abstract: We present a laboratory experiment that examines individuals' willingness to take advice from algorithms provided by an artificial intelligence (AI). We further explored two distinguished channels behind the algorithm, the data and the mechanism. We find that providing data behind the algorithm can substantially reduce individual willingness to take advice from AI (the substitution effect); whereas providing the mechanism does not further crowd out algorithm reliance. If any, participants are slightly more favorable to AI when both data and mechanism are provided compared to when only data is provided (the information disclosure effect). Overall, participants do not take enough advice from the algorithm and the degree of algorithm aversion is sensitive to cognitive capability and self-confidence. Our results shed light on when and why we trust algorithms, which depends on the knowledge we know the algorithm has and the knowledge we have.
Introduction of the speaker: Wei Lijia, Professor and doctoral supervisor of the Economics and Management school of Wuhan University, Dean of the Department of Mathematical Economics and Mathematical Finance, and Executive Director of the Experimental Center for Behavioral Science Research. His research fields include behavioral and experimental economics and applied econometrics. More than 30 papers have been published in international first-class SSCI academic journals such as Marketing Science, Econometric Theory, Journal of Economic Behavior & Organization. And authoritative Chinese academic journals such as China Industrial Economics, China Economic Quarterly, System Engineering Theory and Practice, etc. He has presided over National Natural Science Foundation projects, Humanities and Social Science Foundation of the Ministry of Education and other national and provincial projects.