A systematic review of empirical research on deepfakes

The emergence of deepfakes makes it increasingly challenging to discern the authenticity of sources and messages. Despite the rapid accumulation of research on deepfakes in recent years, detailed reviews of the existing literature that integrate empirical findings in this area and guide future research remain sparse. To this end, this study conducted a systematic review of 163 quantitative and qualitative studies on deepfakes across different disciplines. The results showed that most studies have been conducted in Western countries, with relatively little attention given to Asian countries and even less to African countries. Although there is diversity in the methodological approaches, research on deepfakes is still in its early stages of theory building and testing. Furthermore, extant research focuses primarily on the political domain and topics related to deepfake identification and detection. In light of current evidence, we propose an agenda for future research in this interdisciplinary field.