Nearly three in five students say they turn to AI when sad, stressed, or lonely, JED research finds

Preliminary data from more than 5,500 middle and high school students across 21 U.S. schools was previewed at a bipartisan Capitol Hill briefing, with the full study due in the fall.

The Jed Foundation's bipartisan briefing on Capitol Hill on 21 July drew 150 congressional staff, policymakers, researchers, and mental health professionals. Credit: The Jed Foundation

Four in five middle and high school students have used some form of AI, and nearly three in five have turned to it while feeling sad, stressed, or lonely, according to preliminary findings previewed by The Jed Foundation (JED) at a congressional briefing in Washington, D.C.

The suicide prevention nonprofit, which partners with high schools, colleges, school districts, and community organizations, gathered the data between October 2025 and April 2026 from more than 5,500 students at 21 U.S. schools and school districts already enrolled in its programs. Dr. Kayla Anderson, JED's vice president of research and evaluation, led the work. Full results are expected in the fall.

Academic use dominates, but it is not the whole picture. More than half of students reported using AI for schoolwork. Nearly one third used it for relationship or problem-solving advice, and more than one in ten talked to AI for companionship.

The finding likely to draw the most policy attention concerns what happens next. Students who used AI for emotional or mental health support, for problem-solving advice, during difficult emotions, or for companionship were more likely to report poorer mental health, greater loneliness, and a greater history of suicidal thoughts and behaviors. JED presents these as associations. The organization's summary does not state that AI use causes those outcomes, and the sample is drawn from schools that have opted into JED's mental health programming rather than from a nationally representative panel.

Usage patterns also split by age and geography. High school students and rural students were more likely to use AI at all. Middle school students and urban students were more likely to use it specifically when distressed.

Youth Advisory Council member calls for guardrails and crisis line routing

The briefing, titled "Navigating AI: Youth Perspectives, Mental Health, and Public Policy," drew 150 congressional staff, policymakers, youth advocates, researchers, and mental health professionals on 21 July. It was led by Dr. Zainab Okolo, JED's senior vice president of policy, advocacy, and government relations, with keynote addresses from JED CEO John MacPhee and Dr. Stephanie Haridopolos, principal deputy assistant secretary for health policy and director of national health communications in the office of the Surgeon General.

Okolo frames the generational gap as the reason young people should shape the rules: "Unlike many adults who remember life before constant digital connection, today's young people are growing up with AI as an increasingly intimate part of their social, academic, and emotional lives. Their experiences and perspectives must therefore be central to the policies governing these technologies."

Hanae Noirbent, a member of JED's Youth Advisory Council, argued from the panel that regulation and utility are not mutually exclusive. "We want guardrails implemented that will protect us from the clear risks presented by AI-powered algorithms and generative AI platforms such as chatbots," she says. "We also know that there's a real opportunity for AI to serve as an educational support tool, and even as a tool that directs us to lifesaving resources such as the 988 Suicide and Crisis Lifeline."

Surgeon General's office points to chatbot use in screen time warning

Haridopolos used her address to present her office's Warning on the Harms of Screen Use, which covers emerging concerns about children interacting with chatbots. "We are calling for urgent action to protect children at home, in schools, and across platforms," she states. "Kids are growing up in digital spaces that were never designed for their safety, and these online experiences are shaping how they think, feel, and interact in real life."

The panel also included Allie Maloney, policy analyst at Americans for Responsible Innovation, Blake Hurst, a JED Youth Advisory Council member and student at the University of Notre Dame, and Bhavani Kola, department chair at Tarrant County College and founder and CEO of Eclatech Solutions. Discussion covered state-level responses to the technology and considerations for federal policymaking.

In a LinkedIn post, MacPhee said the briefing was organized by Okolo and Martha Sanchez, JED's director of policy, with Anderson previewing the research.

This was JED's second AI-focused congressional briefing since it published recommendations for policymakers and technology companies. The full research findings are scheduled for release in the fall.