Using ChatGPT to write a literature review on autologous fat grafting.
Access to this document is restricted. Some items have been embargoed at the request of the author, but will be made publicly available after the "No Access Until" date.
During the embargo period, you may request access to the item by clicking the link to the restricted file(s) and completing the request form. If we have contact information for a Cornell author, we will contact the author and request permission to provide access. If we do not have contact information for a Cornell author, or the author denies or does not respond to our inquiry, we will not be able to provide access. For more information, review our policies for restricted content.
BACKGROUND: ChatGPT is a large language model (LLM) that has been proposed as a scientific writing tool, though its ethical use remains a highly debated topic within the academic community. This article defines the strengths and weaknesses of ChatGPT in writing a plastic surgery literature review and describes proper methodologies for optimizing GPT-generated output. METHODS: ChatGPT-4o was prompted to brainstorm topics for a literature review on plastic surgery. Autologous fat grafting was chosen and ChatGPT generated each section of the literature review with citations, which were subsequently evaluated for accuracy. The ability of medical professionals to discriminate between a ChatGPT-generated and published fat grafting abstract was assessed. RESULTS: ChatGPT successfully conceived and performed a literature review on autologous fat grafting. The model performed well in outline creation, article summarization, and editing content. It generated a professional review of fat grafting, though its claims were generalized, not completely factual, and lacked accurate citations. ChatGPT provided 21 citations, 5 of which correctly referenced a real article. Eight contained errors in their publication details, such as publication dates and author lists. The remaining 8 were unable to be found in PubMed (hallucinated). Medical professionals were unable to distinguish ChatGPT-generated material from a published abstract. CONCLUSIONS: With appropriate vigilance, ChatGPT may be cautiously used as a writing assistant throughout the literature review process; however, authors must verify all scientific claims and citations. ChatGPT's greatest limitation remains its tendency to hallucinate, which undermines the reliability of a generated manuscript and perpetuates inaccurate information.