Rethinking University Mental Health for the AI Era: From Crisis Response to Core Mission

Importance Universities have become frontline mental health care settings for adolescents and young adults, a group with rapidly rising rates of common psychiatric disorders. Conventional campus counseling services cannot meet this demand. Students are increasingly turning to unregulated generative artificial intelligence (AI) tools. There is an urgent need for an evidence-informed architecture that integrates digital tools and AI into university mental health systems safely, equitably, and at scale. Objective To synthesize epidemiologic and clinical trial data on student mental health and digital interventions and integrate these findings with experience in institutional governance to propose an integrated, data-driven, stepped-care model for student mental health, implemented and governed at the university level, that begins with universal prevention, tracks need, and uses precision treatment selection to monitor and match students in need with the least-intensive effective level of care. Evidence Review We draw on findings from the World Mental Health International College Student (WMH-ICS) initiative, national surveys of student mental health and service use, randomized clinical trials of student internet-delivered digital interventions, implementation studies of digital navigators and task sharing, and emerging evaluations of AI-supported mental health systems. Findings WMH-ICS studies find that 37% to 58% of first-year students worldwide meet 12-month criteria for a DSM-5 disorder, with low treatment rates and substantial impairment in academic performance. This unmet need exceeds the capacity of traditional short-term, face-to-face counseling services, even with expanded staffing and telehealth contracts. Hybrid stepped-care models combining universal prevention with light-touch human support (eg, digital navigators) and low-intensity digital interventions offer a more scalable approach that can expand reach, optimize use of clinicians’ limited time, and improve outcomes. Building on these strands, we outline a 2-step system anchored in prematriculation life skills training and a guided first-year cohort model, embedded within a broader digital ecosystem that includes measurement, precision treatment algorithms, and explicit AI governance. Conclusions and Relevance Reframing student mental health as core to the mission of universities justifies investment in digitally enabled care. With appropriate guardrails, AI can be integrated as a regulated component of this ecosystem, extending the reach of effective interventions while maintaining safety, equity, and institutional accountability.