Writing the paper “Unveiling artificial intelligence: an insight into ethics and applications in anesthesia” implementing the large language model ChatGPT: a qualitative study

Abstract

Background: This text “Unveiling artificial intelligence: an insight into ethics and applications in anesthesia” addresses the potential applications and limits of artificial intelligence (AI) in anesthesia. The written content was produced with the help of ChatGPT. It is a large language model (LLM) capable of performing a variety of natural language processing (NLP) tasks. The aim of this paper is to evaluate the ability and limits of the chatbot as an aid in writing a scientific text. Methods: In this qualitative study, five domain experts (MC, JM, VB, AV, and EGB) elaborated on the questions for the chatbot and integrated the answers based on the knowledge of the available scientific literature. The output was incorporated, edited, and revised based on the knowledge of the available scientific literature. The first author (MC) assessed the effectiveness of the output and its applicability in writing the paper, as a percentage of words used. All the authors expressed an opinion on the quality of the text and highlighted the pros and cons of using the technology. Results: In the final draft of the text, the scientific output from ChatGPT has been significantly revised and integrated. Many important inaccuracies were found. Conclusions: The utilization of LLMs like ChatGPT presents notable limitations. However, by effectively harnessing specific attributes, these tools can provide advantages for specific tasks. Consequently, it becomes crucial to establish guidelines for their application in scientific writing.