Exploring AI Hallucinations of ChatGPT: Reference Accuracy and Citation Relevance of ChatGPT Models and Training Conditions
Abstract
Introduction Large language model-based generative AI tools, such as the Chat Generative Pre-trained Transformer (ChatGPT) platform, have been used to assist with writing academic manuscripts. Little is known about ChatGPT's ability to accurately cite relevant references in health care simulation-related scholarly manuscripts. In this study, we sought to: (1) determine the reference accuracy and citation relevance among health care simulation debriefing articles generated by 2 different models of ChatGPT and (2) determine if ChatGPT models can be trained with specific prompts to improve reference accuracy and citation relevance. Methods The ChatGPT-4 and ChatGPT o1 models were asked to generate scholarly articles with appropriate references based upon three different article titles about health care simulation debriefing. Five articles with references were generated for each article title-3 ChatGPT-4 training conditions and 2 ChatGPT o1 training conditions. Each article was assessed independently by 2 blinded reviewers for reference accuracy and citation relevance. Results Fifteen articles were generated in total: 9 articles by ChatGPT-4 and 6 articles by ChatGPT o1. A total of 60.4% of the 303 references generated across 5 training conditions were classified as accurate, with no significant difference in reference accuracy between the 5 conditions. A total of 22.2% of the 451 citations were classified as highly relevant, with no significant difference in citation relevance a…
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Studies in radiology, psychiatry, and clinical rehabilitation have reported fabrication rates ranging from 20 to 70%.
“Studies in radiology, psychiatry, and clinical rehabilitation have reported fabrication rates ranging from 20 to 70%”
From How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations (2026) · cites this paper as [14]
Cites
High Rates of Fabricated and Inaccurate References in ChatGPT-Generated Medical Content
Bhattacharyya, M., Miller, V. M., Bhattacharyya, D., & Miller, L. E. (2023) · Cureus · doi:10.7759/cureus.39238 · Cited by 294 · see citing works
Using AI to write scholarly publications
Hosseini, M., Rasmussen, L. M., & Resnik, D. B. (2024) · Accountability in Research · doi:10.1080/08989621.2023.2168535 · Cited by 286 · see citing works
Zielinski, C., Winker, M. A., Aggarwal, R., Ferris, L. E., Heinemann, M., Lapeña Jr, J. F., Pai, S. A., Ing, E., Citrome, L., Alam, M., et al. (2023) · Colombia Medica · doi:10.25100/cm.v54i3.5868 · Cited by 127 · see citing works
The role of using ChatGPT AI in writing medical scientific articles
Benichou, L. (2023) · Journal of Stomatology, Oral and Maxillofacial Surgery · doi:10.1016/j.jormas.2023.101456 · Cited by 73 · see citing works
Zielinski, C., Winker, M. A., Aggarwal, R., Ferris, L. E., Heinemann, M., Lapeña, J. F., Pai, S. A., Ing, E., Citrome, L., Alam, M., et al. (2024) · Current Medical Research and Opinion · doi:10.1080/03007995.2023.2286102 · Cited by 59 · see citing works
Kim, S.-J. (2024) · Science Editing · doi:10.6087/kcse.343 · Cited by 42 · see citing works
Reference accuracy in the general surgery literature
Awrey, J., Inaba, K., Barmparas, G., Recinos, G., Teixeira, P. G. R., Chan, L. S., Talving, P., & Demetriades, D. (2011) · World Journal of Surgery · doi:10.1007/s00268-010-0912-7 · Cited by 33 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Naser, M. (2026) · arXiv · doi:10.48550/arxiv.2603.03299 · Cited by 1 · see citing works
“Studies in radiology, psychiatry, and clinical rehabilitation have reported fabrication rates ranging from 20 to 70%, depending on the model version, prompt specificity, and the availability of relevant literature in the model's training corpus [12 – 14].”
- Authors
- Cheng, A., Nagesh, V., Eller, S., Grant, V., & Lin, Y.
- Affiliations
- University of Calgary (CA)
- Venue
- Simulation in Healthcare: The Journal of the Society for Simulation in Healthcare
- Published
- 2025-08-07
- Volume
- 20
- Issue
- 6
- Pages
- 413-418
- ISSN
- 1559-2332 (ISSN-L)
- 1559-713X
- OpenAlex
- W4413042237
- Cited by
- 16
- References
- 7
- Citations by year
- 2026: 14 · 2025: 2
- FWCI
- 3.82
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Simulation in Healthcare The Journal of the Society for Simulation in Healthcare (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Academic Writing and Publishing
- Social Sciences › Safety Research › Academic integrity and plagiarism
- Keywords
- ChatGPT, AI hallucination, citation accuracy, large language models, ChatGPT-4, academic writing
- MeSH terms
- Simulation Training, Generative Artificial Intelligence, Humans
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.
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
