https://doi.org/10.4081/itjm.2026.2648
Preserving clinical method in artificial intelligence-assisted diagnosis: a cognitive-safety framework against premature diagnostic closure
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.
Published: 21 October 2026
Generative artificial intelligence (AI) can support differential diagnosis, but final-answer accuracy does not by itself establish diagnostic safety. This narrative review examines how AI exposure may alter hypothesis generation and proposes AI-associated differential diagnosis narrowing as a testable process-level construct of the clinician-AI system. The construct describes reduced differential breadth, flexibility, or counterfactual exploration after AI exposure; it is not a validated syndrome and does not imply causation. Anchoring, automation bias, fluent language, and interaction order are possible mechanisms, while premature closure and diagnostic error are potential downstream consequences. We organize candidate safeguards through the GENERATE-Dx mnemonic, which preserves independent hypothesis generation, separates likelihood from danger, makes missing and discordant data explicit, reopens alternatives, and defines verification and reassessment. Candidate endpoints include pre/post differential breadth, retention of expert-adjudicated rare-but-dangerous diagnoses, counterfactual exploration, confidence calibration, management decisions, and reassessment. The framework, metrics, and figures are heuristic proposals. Prospective studies should compare interaction order and interface design and evaluate whether AI supports accurate decisions without weakening the clinical method that makes diagnosis safe.
Downloads
Graber ML, Franklin N, Gordon R. Diagnostic error in internal medicine. Arch Intern Med 2005;165:1493-9. DOI: https://doi.org/10.1001/archinte.165.13.1493
Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med 2003;78:775-80. DOI: https://doi.org/10.1097/00001888-200308000-00003
Committee on Diagnostic Error in Health Care; Board on Health Care Services; Institute of Medicine, et al. Improving diagnosis in health care. Washington, DC: National Academies Press; 2015.
Goddard K, Roudsari A, Wyatt JC. Automation bias: a systematic review of frequency, effect mediators, and mitigators. J Am Med Inform Assoc 2012;19:121-7. DOI: https://doi.org/10.1136/amiajnl-2011-000089
Lyell D, Coiera E. Automation bias and verification complexity: a systematic review. J Am Med Inform Assoc 2017;24:423-31. DOI: https://doi.org/10.1093/jamia/ocw105
Khera R, Simon MA, Ross JS. Automation bias and assistive AI: risk of harm from AI-driven clinical decision support. JAMA 2023;330:2255-7. DOI: https://doi.org/10.1001/jama.2023.22557
Parasuraman R, Riley V. Humans and automation: use, misuse, disuse, abuse. Hum Factors 1997;39:230-53. DOI: https://doi.org/10.1518/001872097778543886
Lee JD, See KA. Trust in automation: designing for appropriate reliance. Hum Factors 2004;46:50-80. DOI: https://doi.org/10.1518/hfes.46.1.50.30392
Goh E, Gallo R, Hom J, et al. Large language model influence on diagnostic reasoning: a randomized clinical trial. JAMA Netw Open 2024;7:e2440969. DOI: https://doi.org/10.1001/jamanetworkopen.2024.40969
McDuff D, Schaekermann M, Tu T, et al. Towards accurate differential diagnosis with large language models. Nature 2025;642:451-7. DOI: https://doi.org/10.1038/s41586-025-08869-4
Cabitza F, Campagner A, Ronzio L, et al. Rams, hounds and white boxes: investigating human-AI collaboration protocols in medical diagnosis. Artif Intell Med 2023;138:102506. DOI: https://doi.org/10.1016/j.artmed.2023.102506
Dhuliawala S, Komeili M, Xu J, et al. Chain-of-verification reduces hallucination in large language models. In: Findings of the Association for Computational Linguistics: ACL 2024. Bangkok: Association for Computational Linguistics; 2024. pp. 3563-78. DOI: https://doi.org/10.18653/v1/2024.findings-acl.212
Wang X, Wei J, Schuurmans D, et al. Self-consistency improves chain of thought reasoning in language models. In: The Eleventh International Conference on Learning Representations; 2023. Available from: https://openreview.net/forum?id=1PL1NIMMrw.
Madaan A, Tandon N, Gupta P, et al. Self-refine: iterative refinement with self-feedback. Adv Neural Inf Process Syst 2023;36:46534-94. DOI: https://doi.org/10.52202/075280-2019
Yao S, Yu D, Zhao J, et al. Tree of thoughts: deliberate problem solving with large language models. Adv Neural Inf Process Syst 2023;36:11809-22. DOI: https://doi.org/10.52202/075280-0517
Du Y, Li S, Torralba A, et al. Improving factuality and reasoning in language models through multiagent debate. Proc Mach Learn Res 2024;235:11733-63.
Savage T, Nayak A, Gallo R, et al. Diagnostic reasoning prompts reveal the potential for large language model interpretability in medicine. NPJ Digit Med 2024;7:20. DOI: https://doi.org/10.1038/s41746-024-01010-1
Hager P, Jungmann F, Holland R, et al. Evaluation and mitigation of the limitations of large language models in clinical decision-making. Nat Med 2024;30:2613-22. DOI: https://doi.org/10.1038/s41591-024-03097-1
Schmidgall S, Harris C, Essien I, et al. Evaluation and mitigation of cognitive biases in medical language models. NPJ Digit Med 2024;7:295. DOI: https://doi.org/10.1038/s41746-024-01283-6
Di Bello F, Angioni D. Evidence-like error and physician responsibility in retrieval-augmented clinical artificial intelligence. AboutOpen 2026;13:22-30. DOI: https://doi.org/10.33393/ao.2026.3822
Mamede S, Schmidt HG, Penaforte JC. Effects of reflective practice on the accuracy of medical diagnoses. Med Educ 2008;42:468-75. DOI: https://doi.org/10.1111/j.1365-2923.2008.03030.x
Mamede S, van Gog T, van den Berge K, et al. Effect of availability bias and reflective reasoning on diagnostic accuracy among internal medicine residents. JAMA 2010;304:1198-203. DOI: https://doi.org/10.1001/jama.2010.1276
Chen SF, Alyakin A, Seas A, et al. LLM-assisted systematic review of large language models in clinical medicine. Nat Med 2026;32:1152-9. DOI: https://doi.org/10.1038/s41591-026-04229-5
Thirunavukarasu AJ, Ting DSJ, Elangovan K, et al. Large language models in medicine. Nat Med 2023;29:1930-40. DOI: https://doi.org/10.1038/s41591-023-02448-8
Tirotta D, Di Bello F, Dentali F, et al. FADOI official position on artificial intelligence in internal medicine. Ital J Med 2026;20:2482. DOI: https://doi.org/10.4081/itjm.2026.2482
Wang G, Zhang K, Jiang J, et al. Human-large language model collaboration in clinical medicine: a systematic review and meta-analysis. NPJ Digit Med 2026;9:195. DOI: https://doi.org/10.1038/s41746-026-02382-2
CRediT authorship contribution
Daniele Angioni: study concept and draft of the manuscript. Fabio Di Bello: conceptual development and critical revision. Daniela Tirotta: critical revision of the clinical content. Andrea Montagnani: supervision of the internal medicine perspective and critical revision of the manuscript. All authors read and approved the final manuscript and agreed to be accountable for all aspects of the work.
How to Cite
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
PAGEPress has chosen to apply the Creative Commons Attribution NonCommercial 4.0 International License (CC BY-NC 4.0) to all manuscripts to be published.