Familiarity With Generative Artificial Intelligence Chatbot Prompt Engineering for Academic Research: A Large-Scale, International Cross-Sectional Survey of Medical Researchers

Le résumé fourni par la source

Generative artificial intelligence chatbots (AICs), such as ChatGPT, Gemini, Meta AI, Claude AI, and Microsoft Co-Pilot, are increasingly used in academic research. Prompt engineering, the practice of crafting effective inputs to elicit high-quality chatbot responses, has emerged as a critical skill. Although AICs are now applied across all stages of the research process, from ideation and design to analysis and dissemination, little is known about how medical researchers understand and use prompt engineering in this context. This study aims to examine medical researchers’ familiarity with, use of, and attitudes toward prompt engineering when using generative AI chatbots throughout the research process. We will conduct a large-scale, international, anonymous cross-sectional survey of corresponding authors of articles published in MEDLINE-indexed journals. The survey will be distributed via email and administered using SurveyMonkey. Eligible participants must self-identify as medical researchers, have a doctoral or master’s level academic degree or significant research experience, and be able to complete the survey in English. The instrument includes closed- and open-ended questions exploring demographic characteristics, patterns of AIC use, perceived benefits and risks, and institutional guidance related to prompt engineering. Descriptive and exploratory statistical analyses will be performed, alongside qualitative thematic analysis of open-ended responses. Findings will offer timely insights into the role of prompt engineering in academic research and inform future institutional policies, training efforts, and research on responsible AIC integration.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.