Technological advances in AI have empowered machines to impersonate humans. The AI agents can learn from our data and imitate our appearance, thoughts, and skills. They could be called assistants, job takers, or toys, but these labels ignore AI's potential impact to disrupt what it means for us to be unique individuals. This talk emphasizes the need to augment the current AI discourse with attention to the basics of individuality, such as selfhood, values, and relationships. To this end, I will interweave recent findings from three design fiction studies. In these studies, my team elicited people's perception of AI's impact on selfhood and relationships by exposing them to speculative scenarios about near-future technologies, such as AI human clones, AI ghostwriters, and AI teachers. Findings from these studies foreshadow looming tensions created by human-like nonhuman beings, including the uncanny valley, distrust, and impression management, and have implications for the ongoing discourse on AI's societal impact in areas such as ethics, justice, and sustainability.

Dongwook Yoon is an Assistant Professor at the Department of Computer Science, University of British Columbia, and a member of Designing for People (DFP) and CAIDA. His overarching goal is to make computer-mediated social interactions richer, more inclusive, and more humane. He attends to problem domains where technology design does not match the social process. To make the technology fit human social interactions, his research assesses user needs in socio-technical systems and addresses such needs by realizing and testing novel design interventions. He earned his Ph.D. from Cornell in 2017 and his M.S. and B.S. from Seoul National University in 2009 and 2007, respectively. His research programs are supported by generous funds from UBC, NSERC, KIST, Adobe, Huawei, and Microsoft.