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

TL;DR

This study investigates non-consensual synthetic intimate imagery (NSII), focusing on AI-generated deepfake pornography, across 10 countries with a sample of 16,693 adults. Using a large online survey and qualitative responses, it documents low familiarity overall but strong public condemnation of NSII, with 2.2% reporting victimization and 1.8% perpetration, and shows men reporting higher familiarity and engagement than women. The findings reveal that most NSII activity is passive (e.g., viewing celebrity deepfakes) and that existing laws are variably effective, underscoring the need for education, platform policies, and technical safeguards (e.g., detection, provenance). The work highlights gender differences in attitudes and experiences and argues for comprehensive interventions—educational programs, policy reforms, and developer/designer strategies—to deter creation and distribution and to support potential victims.

Abstract

Deepfake technologies have become ubiquitous, "democratizing" the ability to manipulate photos and videos. One popular use of deepfake technology is the creation of sexually explicit content, which can then be posted and shared widely on the internet. Drawing on a survey of over 16,000 respondents in 10 different countries, this article examines attitudes and behaviors related to "deepfake pornography" as a specific form of non-consensual synthetic intimate imagery (NSII). Our study found that deepfake pornography behaviors were considered harmful by respondents, despite nascent societal awareness. Regarding the prevalence of deepfake porn victimization and perpetration, 2.2% of all respondents indicated personal victimization, and 1.8% all of respondents indicated perpetration behaviors. Respondents from countries with specific legislation still reported perpetration and victimization experiences, suggesting NSII laws are inadequate to deter perpetration. Approaches to prevent and reduce harms may include digital literacy education, as well as enforced platform policies, practices, and tools which better detect, prevent, and respond to NSII content.

Figures (10)

• Figure 1: Heatmap of Mean Criminalization Attitudes by Country. Scale ranges from -2 ("Definitely should not be a crime") to 2 ("Definitely should be a crime"), with a midpoint of 0 "Not sure"). Brighter red is more deserving of criminalization, darker blue is less deserving of criminalization. Standard errors are presented in parentheses.
• Figure 2: Deepfake Pornography Behaviors by Country. Percentage of all respondents who indicated engaging, or not, in a particular behavior. 95% confidence interval bars presented in grey.
• Figure 3: Percentage of Respondents Reporting Different Types of Deepfake Pornography Video Victimization, by Country. 95% CIs in grey.
• Figure 4: Percentage of Respondents Reporting Different Types of Deepfake Pornography Video Perpetration, by Country. 95% CIs in grey.
• Figure 5: Heatmap of Mean Criminalization Attitudes by Respondent Gender. Scale ranges from -2 ("Definitely should not be a crime") to 2 ("Definitely should be a crime"), with a midpoint of 0 ("Not sure"). Brighter red is more deserving of criminalization, darker blue is less deserving of criminalization. Standard errors are presented in parentheses.