Violation of my body: Perceptions of AI-generated non-consensual (intimate) imagery

View on arXiv ← Back to list

Authors: Natalie Grace Brigham, Miranda Wei, Tadayoshi Kohno, Elissa M. Redmiles

Published: 2024-06-08 16:57:20+00:00

AI Summary

This study surveyed 315 US individuals on their views regarding the non-consensual creation and sharing of AI-generated intimate imagery (AIG-NCII). Respondents strongly opposed the creation and sharing of AIG-NCII, particularly if depicting sexual acts, though seeking out such content was more acceptable to some.

Abstract

AI technology has enabled the creation of deepfakes: hyper-realistic synthetic media. We surveyed 315 individuals in the U.S. on their views regarding the hypothetical non-consensual creation of deepfakes depicting them, including deepfakes portraying sexual acts. Respondents indicated strong opposition to creating and, even more so, sharing non-consensually created synthetic content, especially if that content depicts a sexual act. However, seeking out such content appeared more acceptable to some respondents. Attitudes around acceptability varied further based on the hypothetical creator's relationship to the participant, the respondent's gender and their attitudes towards sexual consent. This study provides initial insight into public perspectives of a growing threat and highlights the need for further research to inform social norms as well as ongoing policy conversations and technical developments in generative AI.

Key findings

Creating, sharing, or seeking AIG-NCII was considered far less acceptable than other forms of non-consensual synthetic media. Intimate partners creating such media were viewed more favorably than strangers, but only when intent was not harmful. Lack of consent was the most common reason for unacceptability.

Approach

The researchers used a vignette-based survey to assess attitudes towards different scenarios involving non-consensual synthetic media. Cumulative link mixed models analyzed responses, considering factors like the depicted action, creator-subject relationship, intent, sexual consent attitudes, and gender.

Datasets

Survey data from 315 US participants recruited through Prolific.

Model(s)

Cumulative link mixed models (CLMMs)

Author countries

USA

← Previous