Evidence-Graded Decision Authorization for Safe Clinical AI: A Constrained Reasoning Framework

Abstract excerpt

Clinical AI systems have achieved strong predictive performance; however, prediction accuracy is not sufficient for clinical safety. Retrieval-augmented generation (RAG) improves factual accuracy, and general-purpose LLM guardrails constrain surface-level output safety, but these mechanisms do not govern the inferential gap between available clinical evidence and permissible clinical claims. We propose Evidence-Gr...

Topics

Open a Topic to create a Post that cites this publication.

• Artificial Intelligence
• Decision Support Systems, Clinical
• Diagnosis, Differential
• Electronic Health Records
• Genetic Variation
• Genetics, Medical
• Phenotype
• Radiomics and Machine Learning in Medical Imaging

Identifiers and source

Literature Corpus work

11012474-9eee-5999-b90f-333cb4b90bb7

DOI

10.64898/2026.05.19.26353565