Login or signup to connect with paper authors and to register for specific Author Connect sessions (if available).

The Deepfake Effect: Drivers of Misjudgment of Authentic Political Audio-Visual Content
Animesh Kumar, Christoph Burtscher, Andreas Eckhardt
Deepfake disinformation is a growing trend in electoral campaigns. However, deepfakes' potentially more pervasive impact is their influence on how we perceive and engage with authentic political audio-visual content. In this study, we examine cognitive functions that may lead to the state of skepticism and even misidentification of authentic political videos upon exposure to deepfakes. To do so, we explore Arnold Zwicky's three cognitive illusions: frequency, in-group, and out-group, that are likely to underlie the individuals' state of skepticism. Preliminary results from the survey of 400 US voters show that voters frequently misjudge authentic political audio-visuals as deepfakes, highlighting a pervasive problem. We also find early evidence supporting the role of both bottom-up and top-down selective attention in frequency illusion but not for group-based illusions.

AuthorConnect Sessions

No sessions scheduled yet