Work

Encountering Deepfakes: A Thematic Analysis of Comments on a Music Video featuring Deepfake Content

Year: 2025

Type: article

Abstract: Much early research on deepfakes has focused on speculating about potential future harms of the technology. Everyday uses and experiences of existing deepfake content are, surprisingly, comparatively understudied. For example, there is little research that seeks to understand how people encounter and respond to deepfake content in today's entertainment media. In this study, we conducted a thematic analysis of user's comments made in response to a Kendrick Lamar YouTube music video that featured deepfake content. In total, we analysed 1467 comments. Our findings are described by six themes; advancing realism of deepfakes, seeing is not believing, predicting dystopian threats, enhancing the experience of the song, strong emotional reactions, and 'a toy for play'. People generally enjoyed the creativity and emotional impact that deepfakes brought to the video but those who were unaware of deepfake technology were left confused and doubting themselves. This study offers a naturalistic view of how deepfakes are encountered in everyday life.

Source: Proceedings of the ACM on Human-Computer Interaction

DOI: 10.1145/3757406

Authors: Didier Ching, John Twomey, Conor Linehan, Gillian Murphy

Institutions: Science Foundation Ireland, University College Cork, Lero

Language: English

Topic: Misinformation and Its Impacts

Subfield: Sociology and Political Science

Field: Social Sciences

Domain: Social Sciences

Open Access status: hybrid

Funders: Science Foundation Ireland, European Regional Development Fund

Award: G1482659561