Understanding Help-Seeking and Help-Giving on Social Media for Image-Based Sexual Abuse

TL;DR

This study investigates how adults seek and receive help for image-based sexual abuse (IBSA) on Reddit, combining a large-scale data pipeline with qualitative analysis. By processing 5.7 million Reddit posts and validating 261 posts across seven IBSA types, the authors map help-seeking patterns, needs, and the types of assistance provided, revealing that informational, technical, and relational support are most common, while institutional help is less frequently sought. The work highlights sociotechnical gaps in connecting victim-survivors with formal care and offers design directions for technology to support prevention, detection, and safer content-sharing practices, including on-device alerts and expanded controls. Overall, the findings emphasize the need for holistic, trauma-informed interventions that integrate social media communities, platforms, and institutions to better prevent IBSA and support recovery. The study demonstrates the potential for AI-assisted triage and guidance to scale empathetic, accurate, and timely help while recognizing the limits of online peer support and the importance of formal resources and safety planning.

Abstract

Image-based sexual abuse (IBSA), like other forms of technology-facilitated abuse, is a growing threat to people's digital safety. Attacks include unwanted solicitations for sexually explicit images, extorting people under threat of leaking their images, or purposefully leaking images to enact revenge or exert control. In this paper, we explore how people seek and receive help for IBSA on social media. Specifically, we identify over 100,000 Reddit posts that engage relationship and advice communities for help related to IBSA. We draw on a stratified sample of 261 posts to qualitatively examine how various types of IBSA unfold, including the mapping of gender, relationship dynamics, and technology involvement to different types of IBSA. We also explore the support needs of victim-survivors experiencing IBSA and how communities help victim-survivors navigate their abuse through technical, emotional, and relationship advice. Finally, we highlight sociotechnical gaps in connecting victim-survivors with important care, regardless of whom they turn to for help.

Figures (4)

• Figure 1: Data processing to identify help-seeking posts on Reddit. We relied on recent advances in LLMs to automatically sift through 5.7M Reddit posts to identify 113K posts likely related to help-seeking for IBSA. We automatically labelled each post with one of seven types of IBSA. Coders randomly pulled a sample of these posts to manually validate the LLM-generated metadata and reach a minimum threshold of 40 posts for each of the seven IBSA types. As a post could mention multiple types of IBSA, our Post dataset includes a total of 261 posts.
• Figure 2: Initial prompt designed to identify posts that contained any discussion of IBSA.
• Figure 3: Categorization prompt designed to identify the most relevant type of IBSA discussed in a Reddit post. IBSA type names and description differ from types shown in Table \ref{['tab:ibsa-categories']} due to experimental iteration with LLM prompts.
• Figure 4: Categorization prompt to determine whether a post discussed a specific type of IBSA (here, nonfinancial sextortion). A Python parser validated the sequence of Yes and No answers necessary for a category label to apply.