Deepfake Detection by Human Crowds, Machines, and Machine-informed Crowds
This addresses the societal challenge of detecting manipulated media, offering insights into human-machine collaboration for deepfake detection, though it is incremental in comparing existing methods.
The study tackled the problem of distinguishing real videos from deepfakes by comparing human observers and a leading computer vision model, finding they had similar accuracy but different error patterns, and that combining human judgment with model predictions improved accuracy, though inaccurate predictions could reduce it.
The recent emergence of machine-manipulated media raises an important societal question: how can we know if a video that we watch is real or fake? In two online studies with 15,016 participants, we present authentic videos and deepfakes and ask participants to identify which is which. We compare the performance of ordinary human observers against the leading computer vision deepfake detection model and find them similarly accurate while making different kinds of mistakes. Together, participants with access to the model's prediction are more accurate than either alone, but inaccurate model predictions often decrease participants' accuracy. To probe the relative strengths and weaknesses of humans and machines as detectors of deepfakes, we examine human and machine performance across video-level features, and we evaluate the impact of pre-registered randomized interventions on deepfake detection. We find that manipulations designed to disrupt visual processing of faces hinder human participants' performance while mostly not affecting the model's performance, suggesting a role for specialized cognitive capacities in explaining human deepfake detection performance.