Deepfake or Real? How Political Alignment and Believability Influence Reporting on Social Media

Wirtschaftsinformatik & Management · Springer Nature · 2026-10-27 · DOI: 10.1007/s10796-026-10833-0

Аннотация

This study investigates whether online users can distinguish political deepfakes from real videos and whether these perceptions shape their intention to report the content. We conducted an online experiment in which participants viewed deepfake and real videos of two presidential candidates during the early stages of the 2024 US presidential election. Results show that political deepfakes are more likely to be reported than real videos because they are perceived as less believable, suggesting that realism alone does not guarantee credibility. Furthermore, users show better discernment and a stronger intention to report a politically aligned video, a pattern consistent with the possibility that political alignment increases cognitive engagement. Our findings provide empirical evidence on how the heuristic realism of AI-generated videos shapes users’ beliefs and reporting intentions. Using the elaboration likelihood model as an interpretive lens, we offer a possible explanation for why political alignment may narrow the believability-reporting gap in polarized digital environments.

Читать у издателя →

Похожие статьи