AI Chatbot Use for Mental Health Advice Rises Sharply Among US Youth, With Key Disparities Identified

Nearly 1 in 5 US adolescents and young adults now seek mental health advice from artificial intelligence (AI) chatbots, according to research published today in JAMA Pediatrics, raising concerns about how these tools are reshaping the treatment ecosystem, largely without clinician or parent awareness.¹

A Persistent Treatment Gap Meets an Accessible Technology

Suicide remains the second leading cause of death among individuals aged 5 to 24 years, with approximately 1 in 5 high schoolers reporting having considered attempting suicide in a 2023 CDC survey.² While rates of depression and anxiety disorders are high among adolescents, 4 in 10 teenagers with a major depressive episode in the past year report receiving no mental health services.¹ AI chatbots have moved quickly to fill this care gap, as a 2025 Pew Research survey found that 64% of teens aged 13 to 17 had ever used an AI chatbot, and 28% use them daily.³

Building on these findings, researchers from RAND, Harvard Medical School, Harvard Pilgrim Health Care Institute, and affiliated institutions designed a study to assess how many young people are turning to these tools for mental health advice, how often, whether they find the interactions helpful, and whether they are disclosing this use to anyone.¹

The cross-sectional survey was administered in November 2025 through RAND's American Life Panel, a nationally representative internet-based panel. Of 1727 individuals aged 12 to 21 years invited to participate, 1009 completed the survey, for a completion rate of 58.4%. The researchers applied survey weights to align the sample to US population demographics, producing a population-weighted estimate representing more than 42.8 million youth.

Respondents were informed that platforms including ChatGPT, Google Gemini, Meta AI, Snap's My AI, and Character.AI qualify as AI chatbots. They were then asked whether they had ever used a chatbot for mental health advice, how frequently, how helpful they found the responses, and whether they had disclosed that use to anyone. All respondents were also asked whether they had spoken with a physician about their mental health in the prior 6 months. The researchers used multivariable logistic regression to examine variation by demographic and geographic subgroup.

Trends, Disparities in Youth AI Chatbot Use for Mental Health

As of 2025, 19.2% of respondents reported having used an AI chatbot for mental health advice, representing an estimated 8.2 million individuals. This compares with 13.1% recorded in a comparable 2024 survey by the same research team, a 6.1% increase in 1 year.

Among those who used chatbots for mental health advice, 42.8% did so at least monthly, whereas 5.8% reported daily or near-daily use. Perceived helpfulness was high, as 91.7% rated the advice as somewhat or very helpful. The researchers cautioned, however, that this may partly reflect AI chatbots' well-documented tendencies toward affirmation rather than accurate clinical guidance.

The subgroup findings highlight important disparities. Girls and young women were twice as likely as boys and young men to report use (adjusted OR [AOR], 2.10; 95% CI, 1.36-3.23). In addition, respondents aged 18 to 21 years were nearly 4 times as likely to report use compared with those aged 12 to 14 (AOR, 3.65; 95% CI, 1.98-6.74). Young people who had already spoken with a physician about their mental health were also more likely to have used a chatbot (AOR, 1.89; 95% CI, 1.18-3.03), suggesting these tools can function alongside, rather than purely in place of, formal care.

Regarding race, Black youth who used chatbots for mental health support were more than 5 times as likely as White youth to do so at least monthly (AOR, 5.45; 95% CI, 1.44-20.66). The researchers noted that this disparity may be due to reduced access to professional services or perceptions of lower clinician responsiveness to their needs.

As for disclosure, 63.3% of users had not told anyone they were using AI chatbots for mental health advice. Among those who did disclose, friends were the most common confidants (28.0%), followed by trusted adults (16.4%), including parents, teachers, or physicians. Without disclosure, clinicians may remain unaware of potentially influential or inaccurate advice their patients are receiving.

Looking Ahead to AI’s Evolving Role in Youth Mental Health

The authors acknowledged several study limitations, including that the sample size and completion rate introduce uncertainty and possible nonresponse bias. Additionally, the survey did not ask respondents to identify which specific chatbots they used or to characterize the nature of their interactions, which limits their ability to analyze the safety, quality, or clinical appropriateness of the advice.

Still, they emphasized that their findings underscore the urgency of understanding and shaping the evolving role of AI chatbots in youth mental health care.

“As these technologies become increasingly integrated into the daily lives of young people, they should be understood as active contributors in the broader ecosystem of psychological interventions,” the authors concluded.

References

1. McBain RK, Cantor JH, Breslau J, et al. AI chatbot use and disclosure for mental health among US adolescents and young adults. JAMA Pediatr. Published online June 01, 2026. doi:10.1001/jamapediatrics.2026.2015
2. Verlenden JV, Fodeman A, Wilkins N, et al. Mental health and suicide risk among high chool students and protective factors—Youth Risk Behavior Survey, US, 2023. MMWR. 2024;73(4): 79-86. doi:10.15585/mmwr.su7304a9
3. Faverio M, Sidoti O. Teens, social media, and AI chatbots 2025. Pew Research Center. December 9, 2025. Accessed June 1, 2026. https://www.pewresearch.org/internet/2025/12/ 09/teens-social-media-and-ai-chatbots-2025/