Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
One Year Later: The Harms Persist, But So Do We!
ABSTRACT
Background:
General-purpose large language models (LLMs) are increasingly used for mental health-related conversations, yet safety guardrails remain inadequate and inconsistent across clinical conditions.
Objective:
This study examines whether the safety guardrails of general-purpose LLMs hold across a broad range of clinical conditions beyond suicide and self-harm, and how different adversarial framings affect them.
Methods:
We evaluated six proprietary LLMs across 16 DSM-5 conditions using four adversarial attack variants, introducing an eight-dimension harm taxonomy and a multi-dimensional evaluation framework.
Results:
Safeguards held reliably only for suicide and self-harm, while conditions such as eating disorders, substance use disorder, and major depressive disorder exhibited failure rates of up to 100%.
Conclusions:
Ethical design and deployment of these LLMs demand clearly defined harm categories across clinical conditions and implementation of safeguards accordingly. Until such safeguards are in place, these models pose significant risks to vulnerable populations, making their growing integration into publicly available settings (e.g., education, open-access LLMs) particularly concerning.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.