Artificial Intelligence and the Restructuring of University Mental Health Systems: Governance, Equity, and the Risk of Algorithmic Stratification (Preprint)

Abstract

<sec> <title>UNSTRUCTURED</title> The global mental health crisis among university students intensified over the past decade and exacerbated by the COVID-19 pandemic has created urgent demand for scalable interventions. Artificial intelligence is increasingly positioned as a transformative solution, offering tools for early risk detection, digital therapeutics, administrative automation, and personalized care pathways. However, rapid integration of AI into university mental health systems introduces substantial ethical, clinical, and equity-related risks: algorithmic bias, privacy vulnerabilities, dehumanization of care, and digital exclusion particularly for institutions in the Global South. This Viewpoint argues that AI is not merely a technological addition but a structural force reshaping the architecture of student mental health systems. Without deliberate governance redesign, AI risks creating algorithmic stratification: a process by which algorithmic systems differentially classify, prioritize, or surveil students based on biased training data, infrastructural disparities, and contextual misalignment. Drawing on interdisciplinary evidence from digital psychiatry, educational technology, and global mental health research, we critically examine AI's dual role as both solution and source of new challenges. We propose the AUGMENT Framework Accessibility without surveillance, User-centered co-design, Governance and auditability, Model transparency, Equity adaptation, Non-replacement of human care, and Tiered integration to guide responsible deployment within stepped-care systems. We call for ethically grounded, context-sensitive innovation that ensures AI enhances rather than fragments or stratifies student mental health support globally. </sec>

Ähnliche Arbeiten