Deepfake Discourse: Lay Reactions to Synthetic Media Across Contexts of Use
Deepfake technologies — including deep learning and generative adversarial networks (GANs) — generate realistic, human-like representations of real or fictional individuals. These are variously labeled as “deepfakes,” “digital humans,” “virtual humans,” or “synthetic avatars.” Public discourse surrounding deepfakes is divided: while some scholars, educators, and journalists highlight their potential benefits, others emphasize the risks associated with their misuse (Halbryt et al., 2022; Westerlund, 2019). Meanwhile, as deepfakes become more common online, they increasingly affect everyday internet users. What do these users think and feel about deepfakes? Does it matter how the technology is used — or what it is called?
This talk presents two studies exploring lay attitudes toward deepfake videos. The first uses discourse analysis to examine public comments on three social media videos: one warning of deepfake dangers, one featuring a celebrity joking, and one demonstrating beauty filters. Despite differing contexts, commenters expressed similar concerns, often catastrophizing and projecting dystopian futures. The second study is a survey experiment investigating psychological mechanisms behind this catastrophizing. Participants viewed clips either before or after reading a catastrophic comment, then reported emotional reactions and attitudes toward synthetic representations — framed as either “deepfakes” or “synthetic avatars.” Preliminary findings suggest that anxiety is socially contagious, and terminology significantly affects acceptance.
I conclude by proposing explanatory mechanisms for deepfake catastrophizing that include availability heuristics and social norms, as well as suggesting strategies for reframing deepfakes for more balanced discourse.
This lecture will also be live streamed via Zoom. You are welcome to join us either in South Hall or online.
For online participants
Online participants must have a Zoom account and be logged in. Sign up for your free account here. If this is your first time using Zoom, please allow a few extra minutes to download and install the browser plugin or mobile app.
Speaker
Susan C. Herring
Susan C. Herring is professor of information science and linguistics and director of the Center for Computer-Mediated Communication at Indiana University, Bloomington. Trained in linguistics at the University of California, Berkeley, she pioneered in the application of sociolinguistic and discourse analysis methods to computer-mediated communication.
She is the creator of the computer-mediated discourse analysis (CMDA) research paradigm, which she and others have used to analyze structural, pragmatic, interactional, and social phenomena in internet communication. Her recent research has focused on multimodal CMC, including communication mediated by telepresence robots, graphical icons, and video face filters. She is currently researching discourse about, and uses of, deepfake technology. She is editor-in-chief of the journal Language@Internet and a past editor of the Journal of Computer-Mediated Communication, the premier journal in her field.
Contact
If you have questions about this event, please contact Kimmy Fox.
