There is a newer version of the record available.

Published July 5, 2026 | Version v0.2.1

Medical AI Failure Atlas / MedFailBench

Authors/Creators

  • 1. Department of Internal Medicine, Kutahya Emet Dr. Fazil Dogan State Hospital, Kutahya, Turkiye

Description

A clinician-built synthetic evaluation resource for medical AI safety wording, missing variable awareness, escalation boundaries, safety gate review, and clinical severity annotation. The resource is intended for model development feedback and research. It is not clinical advice, not a model ranking, and not a clinically validated decision support tool.

Notes

If you use this resource, please cite it as below.

Files

goktugozkanmd/medical-ai-failure-atlas-v0.2.1.zip

Files (2.2 MB)

Name Size
md5:f369dca63f0401a595a3049456fad079
2.2 MB Preview Download

Additional details