cs.AINov 17, 2026

Right Diagnosis, Unsafe Route: Auditing Clinical AI Decision Trajectories and Corrective Oversight for Validity and Potential Harm

Authors: Guanqiao Chen, Wei Jiang, Shu Zhang, Qian Lin, Guanxu Yu, Siran Li, Chunhui Wang, Tianyu Shi

Organizations: New York University · Xi’an Jiaotong University · The University of Hong Kong · Independent Researcher · 

Abstract

A clinical AI system can reach a correct final diagnosis through intermediate decisions that could have harmed the patient, and can repair an error while introducing new risks. Most evaluations score final diagnoses; even stepwise benchmarks, which release a case's evidence over time, score whether systems reach the right diagnosis or gather the right information, not whether the decisions a patient passes through are safe. We make patient safety the target of stepwise evaluation. A trajectory audit replays real admissions checkpoint by checkpoint, lets each system decide with only the evidence released so far, and rates every test, treatment and disposition for its potential harm to the patient and, separately, for clinical validity, against the documented final diagnosis and physician-written disease rubrics yet on the evidence available when the decision was made. An LLM judge scores every decision, and a physician reference standard (triple annotation and senior adjudication of a stratified sample) calibrates the rates. An oversight audit branches each trajectory at its first actionable error and measures, on the same branches, whether corrective feedback repairs the error and whether it introduces new harm. Across 300 cases from a hospital cardiology cohort, MIMIC-IV and the public ASCENT benchmark, with up to seven AI systems per source, 55.2% of hospital trajectories reaching an acceptable final diagnosis passed through a decision with moderate or severe potential harm and 24.8% through a clinically unacceptable one; the harm rates were 76.2% on MIMIC-IV and 39.6% on ASCENT. Structured corrections resolved 73.6% of first errors versus 24.8% after bare rejection, yet new potentially unsafe decisions followed 30.2% of corrections versus 19.7% of rejections. A correct final diagnosis does not certify a safe trajectory, and a successful correction does not certify a safe recovery.

Figures & tables

Appendix figures & tables17 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. A safety-oriented hypothetico-deductive framework for AI-assisted differential diagnosis

    Jul 9, 2026Fan Ma, Mauro Giuffrè, Donald Wright +12Scientific ReasoningMedical Diagnosis

  2. AI Morbidity and Mortality: A Framework for Clinical AI Failure Review

    Aug 31, 2026Paulius Mui, Dean F. Sittig, Steve Labkoff +1AI Risk ManagementAI Safety

  3. Clinician input steers AI toward accurate and harmful recommendations

    Mar 14, 2026Ivan Lopez, Selin S. Everett, Bryan J. Bunning +10Medical LLMsClinical Language Model Evaluation