An explorative assessment of ChatGPT as an aid in medical education: Use it with caution

Abstract

Objective

To explore the use of ChatGPT by educators and students in a medical school setting.

Method

This study used the public version of ChatGPT launched by OpenAI on November 30, 2022 (https://openai.com/blog/chatgpt/). We employed prompts to ask ChatGPT to 1) generate a content outline for a session on the topics of cholesterol, lipoproteins, and hyperlipidemia for medical students; 2) produce a list of learning objectives for the session; and 3) write assessment questions with and without clinical vignettes related to the identified learning objectives. We assessed the responses by ChatGPT for accuracy and reliability to determine the potential of the chatbot as an aid to educators and as a “know-it-all” medical information provider for students.

Results

ChatGPT can function as an aid to educators, but it is not yet suitable as a reliable information resource for educators and medical students.

Conclusion

ChatGPT can be a useful tool to assist medical educators in drafting course and session content outlines and create assessment questions. At the same time, caution must be taken as ChatGPT is prone to providing incorrect information; expert oversight and caution are necessary to ensure the information generated is accurate and beneficial to students. Therefore, it is premature for medical students to use the current version of ChatGPT as a “know-it-all” information provider. In the future, medical educators should work with programming experts to explore and grow the full potential of AI in medical education.

Keywords:

• ChatGPT
• Medical education

Authors’contributions

ZH, FB, and SRT are responsible for conceiving the study and writing the manuscript; ZH and BF were involved in exploring ChatGPT’s potential; AU and AF also explored ChatGPT’s potential and were involved in manuscript writing.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes on contributors

Zhiyong Han

Zhiyong Han, Ph.D., Fortunato Battaglia, Ph.D./M.D., and Stanley R Terlecky, Ph.D. are professors in the Department of Medical Sciences, Hackensack Meridian School of Medicine, Nutley, NJ, USA.

Abinav Udaiyar

Abinav Udaiyar and Allen Fooks are medical students at Hackensack Meridian School of Medicine, Nutley, NJ, USA.