Journal article
Functional Support from AI: Investigating Preferences of Young Adults for Support Types Provided by AI Chatbots Around Sexual Risks and Abuse
International journal of human-computer interaction, pp 1-24
19 May 2026
Abstract
As young people navigate complex online experiences, they may utilize Artificial Intelligence chatbots for support. Despite research on online support, young adults’ chatbot preferences for sexually sensitive topics remain under-explored., We surveyed 18 to 24-year-olds regarding their preferred chatbot support responses within real-world sexual risk scenarios: nonconsensual image sharing, sexual abuse, and peer pressure. Participants were unaware that responses varied across five support types: informational, functional, esteem, network, and emotional. We compared the respondents’ support preferences using the Chi-Squared test and ANOVA. A statistically significant preference for functional support was identified in the sexual abuse and peer pressure scenarios and among those willing to help in the peer pressure scenario. Qualitative analysis of functional support preference focused on actionable guidance. Our findings revealed support preferences and design implications for chatbots. These findings indicate a priority for utility and safety and utility over sentiment. We conclude with trauma-informed design recommendations.
Metrics
1 Record Views
Details
- Title
- Functional Support from AI: Investigating Preferences of Young Adults for Support Types Provided by AI Chatbots Around Sexual Risks and Abuse
- Creators
- Michal Monselise (Corresponding Author) - Drexel UniversityJordyn Young - University of MichiganMax Song - Drexel UniversityBen Brodfuehrer - Drexel UniversityAfsaneh Razi - Drexel University
- Publication Details
- International journal of human-computer interaction, pp 1-24
- Publisher
- Taylor & Francis
- Number of pages
- 24
- Grant note
- U.S. National Science Foundation: 2439312
This research is supported in part by the U.S. National Science Foundation under grant #2439312. Any opinions, findings, and conclusions or recommendations expressed inthis material are those of the authors and do not necessarily reflect the views of theresearch sponsors. We thank all the participants who participated in the research.
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Information Science
- Web of Science ID
- WOS:001771280700001
- Other Identifier
- 991022182863904721