Non-Consensual Synthetic Intimate Imagery: Prevalence, Attitudes, and Knowledge in 10 Countries

Overview

Authors: Rebecca Umbach, Nicola Henry, Gemma Beard, Colleen Berryessa

Publication Date: 13 February 2024

Link: https://arxiv.org/html/2402.01721v2

Keywords: deepfake pornography, deepfakes, AI-generated images, image-based sexual abuse, non-consensual explicit imagery pornographic imagery

Type: Articles/Reports

Summary

The study "Non-Consensual Synthetic Intimate Imagery: Prevalence, Attitudes, and Perceived Harm" examines the global occurrence of deepfake pornography, with data from over 16,000 participants across ten countries. It found that 2.2% of respondents were victims, while 1.8% admitted to creating or sharing such content. Most people agreed that this form of imagery is harmful, but the presence of laws in some regions hasn't effectively reduced its prevalence. The authors emphasize the need for digital literacy education, stronger platform policies, and better enforcement to address the issue.