This page was translated automatically from Turkish. Read the original

This page is an AI-assisted Turkish summary of the study presented at the 26th World Congress of Psychiatry. The study belongs to the presenting authors; please refer to the original source when citing it.

EV0466

Artificial intelligence in psychiatry: clinical applications, diagnostic advances, and ethical issues

ARTIFICIAL INTELLIGENCE IN PSYCHIATRY: CLINICAL APPLICATIONS, DIAGNOSTIC ADVANCES, AND ETHICAL CHALLENGES

Artificial intelligence can enhance psychiatric diagnosis and continuous monitoring, but its clinical value depends on validation, equity, ethical oversight, and ongoing clinician involvement.

The aim is to examine how artificial intelligence (AI) is currently used in psychiatric diagnosis and to discuss its clinical significance, performance, and key challenges for individual care and community mental health. The method is a narrative review based on publications from 2015 to 2025 retrieved through PubMed, Google Scholar, and SciELO. The reviewed literature suggests that AI-based approaches have the potential to support earlier recognition of symptoms, continuous monitoring, and, in some settings, prediction of the course of illness, with some studies reporting diagnostic accuracy between 80% and 85%. Clinical applications include pattern recognition through machine learning in diagnosis; symptom tracking using wearable devices in monitoring; chatbots for treatment support and AI tools to improve screening and accessibility; models that estimate the risk of relapse and predict illness course; and, in public health, the identification of at-risk populations and resource planning. Examples include speech analysis, wearable devices, chatbot-based tools, and digital phenotyping. The discussion highlights a gap between technological development and real-world clinical practice, noting that the growing use of these technologies outside clinical settings raises questions about safety, reliability, self-diagnosis, and the changing physician–patient relationship. The main clinical and ethical challenges are grouped under data variability, limited generalizability, algorithmic bias and digital inequality, privacy, autonomy and accountability, and inappropriate self-management. In conclusion, AI has the potential to enhance psychiatric diagnosis and care, but its role should remain complementary to clinical judgment and be supported by validation, ethical oversight, and clinician involvement.

All WCP 2026 summaries