More and more people try to use artificial intelligence (AI) tools such as ChatGPT (Chat Generative Pretrained Transformer; OpenAI) for scientific writing1. Whether you agree or disagree with the AI tool usage, it is becoming irresistible that ChatGPT can help to generate a lot of research and scientific articles and you might need to use this powerful tool to increase your productivity and the quality of your works. However, there has been sharp disagreement over ChatGPT being listed as author on research papers2-5. As soon as ChatGPT was released as a free-to-use tool in November 2022, at least four articles credit the AI tool as a co-author6-9. The ethical issue on the authorship was raised among the society of scientific article publishing groups and journal editors, researchers, and publishers are now debating the place of such AI tools in the published literature, and whether it’s appropriate to cite the chatbot as an author.
ChatGPT can write academic essays, summarize research papers, and can even answer questions to pass medical exams. It has produced research abstracts that scientists found it hard to spot that a non-human AI had written them. However, it could also make spam, ransomware, and other malicious outputs4. The Chatbot can cite low quality studies containing false numbers, but sound convincing enough to trick human readers. The most worrisome fact is that journal publishers, peer reviewers, and readers of the journal do not have any censoring machinery or screening tools to detect those errors.
Therefore, the major high-impact journal publishers started to announce their policies since January this year especially about the AI usage and declaration in the submitted articles.(Table 1) The journal Science warned researchers in their Editorial article that submitting any manuscripts that have been produced using these tools amounts to scientific misconduct2. Science’s editor-in-chief, Holden Thorp, announced an editorial policy statement that all paper submissions must be the original work of authors, and that content produced by AI is a form of plagiarism. Authors may use the tool only if they have fully disclosed it and Science has approved it. The journal Nature has also introduced similar rules and will not accept any papers listing ChatGPT or any other AI software as authors but hasn’t banned these types of tools completely. Nature mentioned that researchers using large language model (LLM) tools should document this use in the methods or acknowledgements sections3-5. If a paper does not include these sections, the introduction or another appropriate section can be used to document the use of the LLM. Elsevier, which publishes about 2,800 journals, including Cell and the Lancet, has taken a similar stance to the previous two journals, Science and Nature. Its guidelines allow the use of AI tools “to improve the readability and language of the research article, but not to replace key tasks that should be done by the authors, such as interpreting data or drawing scientific conclusions,” said Elsevier’s Andrew Davis, adding that authors must declare if and how they have used AI tools.
Table 1.
The major scientific journal publishers’ policy on usage of artificial intelligence (AI) tool
| Journal publisher | Journal’s policy statement | Date of statement |
|---|---|---|
| AAAS (American Association for the Advancement of Science; publisher of Science) | We would not allow AI to be listed as an author on a paper we published, and use of AI-generated text without proper citation could be considered plagiarism. | 26 January 2023 (Science 2023;379:313)2 |
| Springer Nature (publisher of Nature) | ChatGPT doesn’t meet the standard for authorship. Authors using LLMs (large language models) in any way while developing a paper should document their use in the methods or acknowledgements sections. |
24 January 2023 (Nature 2023;613:612)5 |
| Elsevier (publisher of Cell and Lancet) | The use of AI tools can improve the readability and language of the research article but cannot replace key tasks that should be done by the authors, such as interpreting data or drawing scientific conclusions. AI and AI-assisted tools cannot be credited as an author on published work. | March 2023 (authorship policy updated; https://www.elsevier.com/about/policies/publishingethics) |
Hopefully, there are screening solutions currently being developed by the big publishers as well. Springer Nature is currently developing its own software to detect text generated by AI. Meanwhile Science said it would consider using detection software built by other companies. Although Journal of the Korean Association of Oral and Maxillofacial Surgeons (JKAOMS) has not opened its official policy for AI usage for the manuscript writing, the editorial office recommends that the authors and readers should be careful to use the LLM Chatbot and please clarify any usage of the tool. Furthermore, please do not list AI tools in the author list. JKAOMS will soon update its policy and keep an open mind regarding the future usage of the LLM tools in scientific writing.
Footnotes
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
References
- 1.Huang J, Tan M. The role of ChatGPT in scientific communication: writing better scientific review articles. Am J Cancer Res. 2023;13:1148–54. [PMC free article] [PubMed] [Google Scholar]
- 2.Thorp HH. ChatGPT is fun, but not an author. Science. 2023;379:313. doi: 10.1126/science.adg7879. https://doi.org/10.1126/science.adg7879. [DOI] [PubMed] [Google Scholar]
- 3.Stokel-Walker C. ChatGPT listed as author on research papers: many scientists disapprove. Nature. 2023;613:620–1. doi: 10.1038/d41586-023-00107-z. https://doi.org/10.1038/d41586-023-00107-z. [DOI] [PubMed] [Google Scholar]
- 4.Gaggioli A. Ethics: disclose use of AI in scientific manuscripts. Nature. 2023;614:413. doi: 10.1038/d41586-023-00381-x. https://doi.org/10.1038/d41586-023-00381-x. [DOI] [PubMed] [Google Scholar]
- 5.Tools such as ChatGPT threaten transparent science; here are our ground rules for their use, author. Tools such as ChatGPT threaten transparent science; here are our ground rules for their use. Nature. 2023;613:612. doi: 10.1038/d41586-023-00191-1. https://doi.org/10.1038/d41586-023-00191-1. [DOI] [PubMed] [Google Scholar]
- 6.Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, et al. ChatGPT, author. Performance of ChatGPT on USMLE: potential for ai-assisted medical education using large language models. medRxiv. 2022 doi: 10.1101/2022.12.19.22283643. https://doi.org/10.1101/2022.12.19.22283643. [Epub ahead of print] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.O'Connor S. Open artificial intelligence platforms in nursing education: tools for academic progress or abuse? Nurse Educ Pract. 2023;66:103537. doi: 10.1016/j.nepr.2022.103537. https://doi.org/10.1016/j.nepr.2022.103537. [DOI] [PubMed] [Google Scholar]
- 8.ChatGPT Generative Pre-trained Transformer; Zhavoronkov A, author. Rapamycin in the context of Pascal's Wager: generative pre-trained transformer perspective. Oncoscience. 2022;9:82–4. doi: 10.18632/oncoscience.571. https://doi.org/10.18632/oncoscience.571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Osmanovic-Thunström A, Steingrimsson S GPT Generative Pretrained Transformer, author. an GPT-3 write an academic paper on itself, with minimal human input? HAL. https://hal.science/hal-03701250. [Epub ahead of print] [Google Scholar]