A paper coauthored by PhD student Zhixuan (Kyrie) Zhou and Mengyi Wei, a PhD student at the Technical University of Munich, Germany, received the Best Paper Award at the 56th Hawaii International Conference on System Sciences (HICSS-56). The conference, which was held in Maui on January 3-6, is one of the longest-standing working scientific conferences in information technology management.
In their paper, "AI Ethics Issues in Real World: Evidence from AI Incident Database," Zhou and Wei discuss how they used the AI Incident Database to investigate how AI ethics incidents occur in the real world as well as their social impact. For their study, they applied a qualitative content analysis to 150 AI incidents, including their time, location, and corresponding news articles. The researchers identified areas that often involve ethical issues of AI, such as intelligent service robots, language/vision models, and autonomous driving, along with areas of AI risk, including physical safety, bad performance, and racial and gender discrimination.
"By picturing the landscape of real-world AI ethics incidents, and inspecting AI risks associated with each application area, we aim to provide a perspective for policy makers to formulate more operable AI guidelines and regulatory frameworks," said Zhou. "Our next steps are to expand our current analysis of AI ethics incidents to broader Twitter discussions, to gain quantitative insights using text mining techniques, and to refine current AI guidelines and regulatory frameworks."
Zhou's research interests are in human-computer interaction and usable security. He combines qualitative and quantitative methods to approach gender issues (e.g., stereotypes) and AI ethics issues (e.g., fake news). He also conducts ethnographic research to understand livestreaming and blockchain practices. Zhou earned his BEng in computer science from Wuhan University.