The DOI matches its Crossref record; 31 of the 31 references it cites were found in a registry or an index; 5 of the 14 sources its claims cite were checked for support.
Automation Bias in Large Language Model Assisted Diagnostic Reasoning Among AI-Trained Physicians
Open accessAbstract
Whether AI-trained physicians are vulnerable to automation bias when LLM use is voluntary remains unknown. Erroneous LLM recommendations significantly degrade performance...
Overview
CoChat Summary
TL;DR
In a single-blind randomized trial, 44 AI-trained Pakistani physicians diagnosed six clinical vignettes with voluntary access to ChatGPT-4o recommendations; those offered deliberately flawed advice in 3 of 6 vignettes scored substantially lower on diagnostic reasoning than those given error-free advice. The authors conclude that erroneous LLM output induces automation bias even among physicians with dedicated AI-literacy training.
Key findings
- Primary outcome (diagnostic reasoning accuracy): control 84.9% (SD 19.7%) vs treatment 73.3% (SD 30.5%); adjusted mean difference −14.0 percentage points (95% CI −18.9 to −9.1; P < .0001). (The abstract reports the same point estimate with a CI of −8.3 to −19.7.)
- Secondary outcome (top-choice diagnosis accuracy per case): 90.5% (SD 28.9) control vs 76.1% (SD 42.5) treatment; adjusted difference −18.3 percentage points (95% CI −26.6 to −10.0; P < .0001).
- AI consultation was voluntary and similar across arms: 68.9% treatment vs 66.7% control (P = .69).
- Subgroup analyses: larger declines among physicians at/above the median 10 years of practice (−16.6 pp, 95% CI −23.1 to −10.1; P < .0001) vs −9.1 pp (95% CI −18.1 to −0.1; P = .0474) as reported; frequent LLM users (≥ weekly) −11.0 pp (95% CI −18.5 to −3.6; P = .0037) vs −10.7 pp among less frequent users (95% CI −24.5 to 3.1; P = .1285); male physicians −25.8 pp (95% CI −33.8 to −17.7; P < .0001) vs −2.1 pp among female physicians (95% CI −9.8 to 5.5; P = .5839).
- Grading reliability: Krippendorff's α = 0.93 across three blinded graders; Cronbach's α = 0.80.
Methods
- Single-blind RCT (ClinicalTrials.gov NCT06963957), conducted June 20–August 15, 2025, at Lahore University of Management Sciences and via video conferencing; 44 physicians (22 per arm) registered with the Pakistan Medical and Dental Council, holding MBBS degrees, all having completed a 20-hour AI-literacy training. Participants were blinded to the study aims; randomization was known only to the study administrator.
- Participants diagnosed 6 vignettes (internal medicine, cardiology, neurology, pediatrics, infectious disease, emergency medicine) in 75 minutes, with voluntary opt-in access to pre-generated ChatGPT-4o recommendations; control received unmodified output, treatment received clinically significant embedded errors in 3 of 6 vignettes. Google "AI Overviews" were blocked via browser extension; conventional online medical resources were permitted. 264 cases completed (132 per arm).
- Primary outcome scored by three blinded physicians using an expert-validated rubric (differential diagnosis accuracy, supporting/opposing evidence, recommended next steps); analysis by intention-to-treat at the case level using linear mixed-effects models with random effects for participant and case; target sample was 50 (25 per arm), with post-hoc power ≥80% given the observed 14.0 pp effect.
Limitations
- Stated by the authors: vignettes may not capture real-world complexity (context, time pressure, multimorbidity); errors were physician-introduced and may be less subtle than real AI errors; findings are limited to ChatGPT-4o; the single-session design cannot show whether bias persists or fades with repeated use; mitigation strategies were not explored; broader healthcare professionals (nurses, physician assistants) were not studied.
- Evident from the text: enrollment fell short of the 50-participant target (44 enrolled, 88% of target); the subgroup analyses were not adjusted for multiple comparisons; the subgroup reporting is internally inconsistent, describing the −16.6 pp group as "at or above the median years of practice" while calling the −9.1 pp group "more experienced physicians"; and the trial was conducted at a single country's institutions with a preprint that has not been peer reviewed.
Model-written summary of a shortened version of the paper’s full text (it is longer than we read). Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
3 Valid · 2 Not valid · 3 Not checked · 31 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“when leading LLMs were tested with 300 physician-validated vignettes containing even one fabricated detail, hallucination rates reached 50-82%.16”
What the sources say
- Supportedcites [16] source
The Methods (page unavailable) describe 300 physician-validated simulated vignettes, each with one fabricated detail. The Results (page unavailable) report default-setting hallucination rates from 50.0% to 82.7% across the tested models and case formats; the claim’s 50–82% is a slightly truncated approximation of that range. (LLM, conf=0.92)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This trial demonstrates that erroneous LLM recommendations significantly degrade physicians' diagnostic performance by inducing automation bias, even in AI-trained physicians.”
boostedstates a causen = 22- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 3 assumptions not shown here.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These analyses indicate that vulnerability to AI-induced diagnostic errors may be concentrated among experienced, male physicians with frequent LLM exposure; a demographic warranting targeted interventions.”
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This finding is particularly notable because physicians were not required to use the LLM; their voluntary consultation and subsequent reliance on erroneous recommendations suggests that technological assistance can override, rather than augment, clinical reasoning, even among users with specialized AI-training.”
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 3 assumptions not shown here.
In the paper
“Emerging evidence demonstrates that physicians are willing to modify their clinical decisions based on LLM assistance.”
What the sources say
- Not matchedcites Goh et al.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
7 more claims
Create a free account to see all 39 claims and to verify the 2 references still open.
- It may state two separate claims as one. 17
- Probably not a substantial claim. 18
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Results
“compared to a 9.1 percentage points reduction among more experienced physicians”
§ Discussion
“Physicians with above-median clinical experience showed a greater decline in accuracy than their less experienced peers (-16.6 vs -9.1 percentage points)”
Citations
Cites
User Acceptance of Information Technology: Toward A Unified View1
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003) · MIS Quarterly · doi:10.2307/30036540 · Cited by 45,035 · cited 1×
“30,31 Furthermore, frequent LLM users (weekly or more) showed significant performance drop (-11.0 percentage points), unlike infrequent users.”
InfluentialComplacency and Bias in Human Use of Automation: An Attentional Integration
Parasuraman, R., & Manzey, D. H. (2010) · Human Factors: The Journal of the Human Factors and Ergonomics Society · doi:10.1177/0018720810376055 · Cited by 1,710
Improving Diagnosis in Health Care
Committee on Diagnostic Error in Health Care, Board on Health Care Services, Institute of Medicine, & The National Academies of Sciences, E. (2015) · doi:10.17226/21794 · Cited by 1,376
Automation bias: a systematic review of frequency, effect mediators, and mitigators
Goddard, K., Roudsari, A., & Wyatt, J. C. (2012) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2011-000089 · Cited by 1,162
Lekadir, K., Frangi, A. F., Porras, A. R., Glocker, B., Cintas, C., Langlotz, C. P., Weicken, E., Asselbergs, F. W., Prior, F., Collins, G. S., et al. (2025) · BMJ · doi:10.1136/bmj-2024-081554 · Cited by 957
Evaluation and mitigation of the limitations of large language models in clinical decision-making
Hager, P., Jungmann, F., Holland, R., Bhagat, K., Hubrecht, I., Knauer, M., Vielhauer, J., Makowski, M., Braren, R., Kaissis, G., et al. (2024) · Nature Medicine · doi:10.1038/s41591-024-03097-1 · Cited by 798
Large Language Model Influence on Diagnostic Reasoning: A Randomized Clinical Trial
Goh, E., Gallo, R., Hom, J., Strong, E., Weng, Y., Kerman, H., Cool, J. A., Kanjee, Z., Parsons, A. S., Ahuja, N., et al. (2024) · JAMA Network Open · doi:10.1001/jamanetworkopen.2024.40969 · Cited by 724
Changes in Rates of Autopsy-Detected Diagnostic Errors Over Time: A Systematic Review
Shojania, K. G., Burton, E. C., McDonald, K. M., & Goldman, L. (2003) · JAMA · doi:10.1001/jama.289.21.2849 · Cited by 703
Types and Origins of Diagnostic Errors in Primary Care Settings
Singh, H., Giardina, T. D., Meyer, A. N. D., Forjuoh, S. N., Reis, M. D., & Thomas, E. J. (2013) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2013.2777 · Cited by 559
Dratsch, T., Chen, X., Rezazade Mehrizi, M., Kloeckner, R., Mähringer-Kunz, A., Püsken, M., Baeßler, B., Sauer, S., Maintz, D., & Pinto dos Santos, D. (2023) · Radiology · doi:10.1148/radiol.222176 · Cited by 369
Research works printed but not linked: 2
- American Medical Association. (2025, May 8). AI in medical education. https://www.ama-assn.org/education/changemeded-initiative/ai-medical-education
- American Medical Association. (2024, February 26). Future of health: The emerging landscape of augmented intelligence (AI) in health care. American Medical Association. Retrieved September 3, 2025, from https://www.ama-assn.org/system/files/future-health-augmented-intelligence-health-care.pdf
Cited by
Cognitive Atrophy Paradox of AI–Human Interaction: From Cognitive Growth and Atrophy to Balance
Kabashkin, I. (2025) · Information · doi:10.3390/info16111009 · Cited by 34
van Zyl, L. E. (2026) · AI & Society · doi:10.1007/s00146-025-02814-9 · Cited by 16
“Here, psychologists may defer to algorithmic recommendations even when confronted with contradictory clinical evidence (Qazi et al. 2025) .”
The Effect of LLM Assistance on Diagnostic Accuracy: A Systematic Review and Meta-AnalysisPreprint
Tretow, I., Ergon, E., Schwebel, M., Treffers, T., Welpe, I. M., & Feuerriegel, S. (2025) · medRxiv · doi:10.64898/2025.12.11.25341476 · Cited by 8
The Deskilling Effect: Is Artificial Intelligence Eroding Clinical Competence?
Topaz, M., Peltonen, L. M., Zhang, Z., & Backonja, U. (2026) · Annals of Internal Medicine · doi:10.7326/annals-26-00613 · Cited by 8
Arslan, İ. B., Kumandaş, M. T., Arslan, D., Aslan, K., Saltoğlu, T., Çalışkan, S. A., & Terzi, M. (2026) · Acta Neurologica Belgica · doi:10.1007/s13760-026-03102-z · Cited by 3
Metadata
- Authors
- Qazi, I. A., Ali, A., Khawaja, A. U., Akhtar, M. J., Sheikh, A. Z., & Alizai, M. H.
- Affiliations
- Lahore University of Management Sciences (PK)
- King Edward Medical University (PK)
- Lahore General Hospital (PK)
- Corresponding author
- Qazi, Ihsan Ayyub
- Venue
- medRxiv
- Publisher
- openRxiv
- Published
- 2025-08-26
- OpenAlex
- W4413640513
- Cited by
- 16
- References
- 31
- Citations by year
- 2026: 14 · 2025: 2
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Licence
- Locations
- medRxiv (acceptedVersion, cc-by-nc-nd)
- Funders
- Lahore University of Management Sciences
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Family Practice › Clinical Reasoning and Diagnostic Skills
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- automation bias, large language models, AI-assisted diagnosis, GPT-4o, human-AI collaboration, AI literacy training, differential diagnosis, patient safety, human oversight of AI, clinical vignettes, prompt engineering, critical evaluation of AI outputs, Pakistan, South Asia
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-07 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Automation Bias in Large Language Model Assisted Diagnostic Reasoning Among AI-Trained Physicians. https://app.cochat.ai/papers/10.1101/2025.08.23.25334280
- The paper
- Qazi, I. A., Ali, A., Khawaja, A. U., Akhtar, M. J., Sheikh, A. Z., & Alizai, M. H. (2025). Automation Bias in Large Language Model Assisted Diagnostic Reasoning Among AI-Trained Physicians. medRxiv. https://doi.org/10.1101/2025.08.23.25334280
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.
