Accessibility settings

Published on in Vol 13 (2026)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/88435, first published .
Man thoughtfully looking at AI brain graphic on computer screen

Between Help and Harm: An Evaluation Study of Mental Health Crisis Handling by Large Language Models

Between Help and Harm: An Evaluation Study of Mental Health Crisis Handling by Large Language Models

Adrian Arnaiz-Rodriguez   1 , PhD ;   Miguel Baidal   1 , MSc ;   Erik Derner   1, 2 , PhD ;   Jenn Layton Annable   3 , MSc ;   Mark Ball   4 , MA, PGCert HE ;   Mark Ince   5 , BPhil, DipSW ;   Elvira Perez Vallejos   6 , PhD ;   Nuria Oliver   1 , PhD

1 ELLIS Alicante, Alicante, Spain

2 Czech Technical University in Prague, Prague, Czech Republic

3 School of Health Science, University of Nottingham, Nottingham, United Kingdom

4 Public Health and Social Care, School of Psychology, University of Derby, Derby, United Kingdom

5 Mental Health Practitioner, Independent scholar, Derby, United Kingdom

6 School of Computer Science & School of Medicine, University of Nottingham, Nottingham, United Kingdom

Corresponding Author:

  • Adrian Arnaiz-Rodriguez, PhD
  • ELLIS Alicante
  • Muelle de Poniente 5, Distrito Digital 5 - Edificio A, Puerto de Alicante
  • Alicante 03001
  • Spain
  • Phone: 34 659540233
  • Email: adrian@ellisalicante.org