" He made me feel less alone": Adolescents’ Engagement with GenAI for Emotional Support.
Abstract
An increasing number of adolescents worldwide experience mental health difficulties, yet access to formal and informal support remains limited. Concurrently, Generative Artificial Intelligence (GenAI) tools have rapidly entered everyday life and become widely accessible to most adolescents. This study examined adolescents’ spontaneous use of GenAI for emotional support, as well as their experiences and level of knowledge regarding GenAI. Participants were 594 Israeli adolescents aged 12 to 18 who completed a survey with multiple-choice and open-ended questions. Results indicate 84% (n=498) use GenAI for emotional support purposes, while 93% (n=552) use it for school-learning purposes. Content analysis identified two primary distress contexts that prompted adolescents to turn to GenAI for support: (1) intrapersonal difficulties including emotional difficulties or psychopathological symptoms (e.g., anxiety, depressive symptoms) and (2) interpersonal difficulties (e.g., conflicts with friends, parents, teachers, or romantic partners). Among adolescents who used GenAI tools for emotional support, 30% reported it to be very helpful, 42% reported feeling substantially understood, and 22% reported a marked reduction in loneliness following usage. Notably, girls reported significantly higher use of GenAI for emotional support and more positive experiences than boys. Many participants noted that the dialogue itself felt relieving, as it gave them a sense that "someone" was there with them. Finally, adolescents demonstrated limited levels of knowledge regarding how the technology underlying GenAI tools operates. The findings demonstrate the daily use of GenAI for emotional support by adolescents and underscore the need for age appropriate GenAI literacy.
Fields of Study & Subjects
Digital Mental Health Interventions
Impact of Technology on Adolescents
Artificial Intelligence in Education
Publication details
- Journal / source
- PsyArXiv (OSF Preprints)
- Publication date
- 12 March 2026
- Language
- EN
- Document type
- Preprint