Differentiating ChatGPT-Generated and Human-Written Medical Texts: Quantitative Study
Open accessThe DOI matches its Crossref record; 18 of the 18 references it cites were found in a registry or an index.
Abstract
Background: Large language models, such as ChatGPT, are capable of generating grammatically perfect and human-like text content, and a large number of ChatGPT-generated texts have appeared on the internet. However, medical texts, such as clinical notes and diagnoses, require rigorous validation, and erroneous medical content generated by ChatGPT could potentially lead to disinformation that poses significant harm to health care and the general public. Objective: This study is among the first on responsible artificial intelligence–generated content in medicine. We focus on analyzing the differences between medical texts written by human experts and those generated by ChatGPT and designing machine learning workflows to effectively detect and differentiate medical texts generated by ChatGPT. Methods: We first constructed a suite of data sets containing medical texts written by human experts and generated by ChatGPT. We analyzed the linguistic features of these 2 types of content and uncovered differences in vocabulary, parts-of-speech, dependency, sentiment, perplexity, and other aspects. Finally, we designed and implemented machine learning methods to detect medical text generated by ChatGPT. The data and code used in this paper are published on GitHub. Results: Medical texts written by humans were more concrete, more diverse, and typically contained more useful information, while medical texts generated by ChatGPT paid more attention to fluency and logic and usually expressed general terminologies rather than effective information specific to the context of the problem. A bidirectional encoder representations from transformers–based model effectively detected medical texts generated by ChatGPT, and the F1 score exceeded 95%. Conclusions: Although text generated by ChatGPT is grammatically perfect and human-like, the linguistic characteristics of generated medical texts were different from those written by human experts. Medical text generated by ChatGPT could be effectively detected by the proposed machine learning algorithms. This study provides a pathway toward trustworthy and accountable use of large language models in medicine.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2019) · arXiv · doi:10.48550/arxiv.1910.13461 · Cited by 8,816
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C. H., & Kang, J. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btz682 · Cited by 7,218
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., & Soricut, R. (2019) · arXiv · doi:10.48550/arxiv.1909.11942 · Cited by 4,308 · cited 2×
“, BERT [2], Roberta [18], Albert [19] ), decoder-based models (eg: GPT [1], GPT2 [20]), encoder-decoder-based models (e.”
Training language models to follow instructions with human feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.02155 · Cited by 4,208
Gilson, A., Safranek, C. W., Huang, T., Socrates, V., Chi, L., Taylor, R. A., & Chartash, D. (2023) · JMIR Medical Education · doi:10.2196/45312 · Cited by 2,078
Alsentzer, E., Murphy, J., Boag, W., Weng, W.-H., Jindi, D., Naumann, T., & McDermott, M. (2019) · Proceedings of the 2nd Clinical Natural Language Processing Workshop · doi:10.18653/v1/w19-1909 · Cited by 1,736
Summary of ChatGPT-Related Research and Perspective Towards the Future of Large Language Models
Liu, Y., Han, T., Ma, S., Zhang, J., Yang, Y., Tian, J., He, H., Li, A., He, M., Liu, Z., et al. (2023) · arXiv · doi:10.48550/arxiv.2304.01852 · Cited by 1,172
Self-citationBioGPT: generative pre-trained transformer for biomedical text generation and mining
Luo, R., Sun, L., Xia, Y., Qin, T., Zhang, S., Poon, H., & Liu, T.-Y. (2022) · Briefings in Bioinformatics · doi:10.1093/bib/bbac409 · Cited by 1,150
Self-citation
Cited by
Summary of ChatGPT-Related Research and Perspective Towards the Future of Large Language Models
Liu, Y., Han, T., Ma, S., Zhang, J., Yang, Y., Tian, J., He, H., Li, A., He, M., Liu, Z., et al. (2023) · arXiv · doi:10.48550/arxiv.2304.01852 · Cited by 1,172
Self-citationSummary of ChatGPT-Related research and perspective towards the future of large language models
Liu, Y., Han, T., Ma, S., Zhang, J., Yang, Y., Tian, J., He, H., Li, A., He, M., Liu, Z., et al. (2023) · Meta-Radiology · doi:10.1016/j.metrad.2023.100017 · Cited by 855
“Secondly, ChatGPT lacks the specialized medical knowledge and training necessary for precise interpretation and analysis of medical images, potentially leading to dangerous misunderstandings or inaccurate diagnoses [52].”
BackgroundSelf-citationLarge language models in healthcare and medical domain: A review
Nazi, Z. A., & Peng, W. (2024) · Informatics · doi:10.3390/informatics11030057 · Cited by 415
A Survey of GPT-3 Family Large Language Models Including ChatGPT and GPT-4
Kalyan, K. S. (2024) · Natural Language Processing Journal · doi:10.1016/j.nlp.2023.100048 · Cited by 335
Younis, H. A., Eisa, T. A. E., Nasser, M., Sahib, T. M., Noor, A. A., Alyasiri, O. M., Salisu, S., Hayder, I. M., & Younis, H. A. (2024) · Diagnostics · doi:10.3390/diagnostics14010109 · Cited by 242
A Survey on LLM-Generated Text Detection: Necessity, Methods, and Future Directions
Wu, J., Yang, S., Zhan, R., Yuan, Y., Wong, D. F., & Chao, L. S. (2023) · arXiv · doi:10.48550/arxiv.2310.14724 · Cited by 212
“Recent research [43, 164] has elucidated distinct disparities between LLM-generated text and human-written text, underscoring a tangible and discernible gap between the two.”
BackgroundA Survey on LLM-Generated Text Detection: Necessity, Methods, and Future Directions
Wu, J., Yang, S., Zhan, R., Yuan, Y., Chao, L. S., & Wong, D. F. (2025) · Computational Linguistics · doi:10.1162/coli_a_00549 · Cited by 212
Oniani, D., Hilsman, J., Peng, Y., Poropatich, R. K., Pamplin, J. C., Legault, G. L., & Wang, Y. (2023) · npj Digital Medicine · doi:10.1038/s41746-023-00965-x · Cited by 143
Contrasting Linguistic Patterns in Human and LLM-Generated News Text
Muñoz-Ortiz, A., Gómez-Rodríguez, C., & Vilares, D. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10903-2 · Cited by 94
Impact of a Digital Scribe System on Clinical Documentation Time and Quality: Usability Study
van Buchem, M. M., Kant, I. M. J., King, L., Kazmaier, J., Steyerberg, E. W., & Bauer, M. P. (2024) · JMIR AI · doi:10.2196/60020 · Cited by 70
“Two previous studies compared human and ChatGPT-written medical texts and reported similar results [22,23].”
ResultPudasaini, S., Miralles-Pechuán, L., Lillis, D., & Llorens Salvador, M. (2025) · Journal of Academic Ethics · doi:10.1007/s10805-024-09576-x · Cited by 70
Radiology-GPT: A Large Language Model for Radiology
Liu, Z., Zhong, A., Li, Y., Yang, L., Ju, C., Wu, Z., Ma, C., Shu, P., Chen, C., Kim, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2306.08666 · Cited by 59
Self-citationKim, S.-J. (2024) · Science Editing · doi:10.6087/kcse.343 · Cited by 42
Fight Fire with Fire: How Much Can We Trust ChatGPT on Source Code-Related Tasks?
Yu, X., Liu, L., Hu, X., Keung, J. W., Liu, J., & Xia, X. (2024) · arXiv · doi:10.48550/arxiv.2405.12641 · Cited by 29
Can we trust academic AI detective? Accuracy and limitations of AI-output detectors
Erol, G., Ergen, A., Gülşen Erol, B., Kaya Ergen, Ş., Bora, T. S., Çölgeçen, A. D., Araz, B., Şahin, C., Bostancı, G., Kılıç, İ., et al. (2025) · Acta Neurochirurgica · doi:10.1007/s00701-025-06622-4 · Cited by 28
Arbel, Y., Gimmon, Y., & Shmueli, L. (2025) · Physical Therapy · doi:10.1093/ptj/pzaf010 · Cited by 21
Kim, N., Homer, M., & Jang, H. (2025) · American Journal of Speech-Language Pathology · doi:10.1044/2025_ajslp-24-00464 · Cited by 9
Metadata
- Authors
- Liao, W., Liu, Z., Dai, H., Xu, S., Wu, Z., Zhang, Y., Huang, X., Zhu, D., Cai, H., Li, Q., Liu, T., & Li, X.
- Venue
- JMIR Medical Education
- Publisher
- JMIR Publications Inc.
- Published
- 2023-12-28
- Volume
- 9
- Pages
- e48904
- ISSN
- 2369-3762 (ISSN-L)
- OpenAlex
- W4390347092
- Cited by
- 116
- References
- 18
- FWCI
- 10.19
- Type
- journal article
- Language
- en
- Open access
- gold
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
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.