Related Experiment Videos
Generative AI in Youth Mental Health Apps: Rapid Review
Helene Høgsdal1, Henriette Kyrrestad1, Sabine Kaiser1
1UiT The Arctic University of Norway, Faculty of Health Sciences, Regional Centre for Child and Youth Mental Health and Child Welfare - North, Tromsø, Norway.
Background:
Mental health apps are frequently used as platforms for delivering digital health interventions to young people. New technology such as generative AI enables a wider range of engaging and more personalized features that can be included in mental health apps. However, there is limited insight into the opportunities for integrating generative AI into such apps.
Objective:
This rapid review aimed to provide a systematic overview of the possibilities of integrating generative AI into mental health apps for young people. Furthermore, the review aimed to report potential benefits and disadvantages of the integration of generative AI into such apps.
Methods:
A systematic search was conducted in 4 databases (PsycInfo, Embase, MEDLINE, and CINAHL) in November 2025 to identify studies that evaluated mental health apps with integrated generative AI. Eligible studies included primary research published between 2022 and 2025, involving a sample of young people aged 11 to 24 years, and written in either English or a Scandinavian language.
Results:
A total of 5 articles were included in the review. The most common method of integrating generative AI into mental health apps for young people was through chatbots. Overall, young people rated the apps as having good usability and quality. Some studies also provided data on effectiveness, showing promising results for outcomes such as depression, anxiety, and distress. None of the studies systematically assessed harmful effects using standardized methods, nor did they report any adverse outcomes related to mental health.
Conclusions:
Young people are generally positive about apps that include generative AI. There is some evidence suggesting that such tools may contribute to a preventive or health-promoting effect on young people's mental health. However, the existing research is limited and characterized by methodological constraints. The lack of reported adverse mental health outcomes might reflect a lack of investigation rather than evidence of no harm. Further research should explore the potential short- and long-term effects of integrating generative AI into mental health apps for young people, as well as systematically mapping possible adverse events.
