From AI to the Table: A Systematic Review of ChatGPT’s Potential and Performance in Meal Planning and Dietary Recommendations
Peiqi Guo, Guancheng Liu, Xiaoling Xiang, Ruopeng An
Abstract
A balanced diet is crucial for preventing diseases and managing existing health conditions. ChatGPT as garnered attention from researchers, including nutrition scientists and dietitians, as an innovative tool for personalized meal planning and dietary recommendations. Objectives: The purpose of this study was to review scientific evidence on ChatGPT’s performance in providing personalized meal plans and generating dietary recommendations. Methods: This systematic review was conducted following the PRISMA guidelines. Keyword-based database searches were performed on PubMed, Web of Science, EBSCO, and Embase. Inclusion criteria included (1) empirical studies and (2) primary research on ChatGPT’s performance in personalized meal planning and dietary recommendations. Results: Twenty-three studies met the inclusion criteria, comprising fourteen validation studies, five comparative studies, and four qualitative studies. Most studies reported that ChatGPT achieved satisfactory accuracy and was often indistinguishable from human dietitians. One study even reported that ChatGPT outperformed human dietitians. However, limitations and risks, such as safety concerns and a lack of real-world implementation, were also identified. Conclusions: ChatGPT shows promise as a relatively reliable innovative tool for personalized meal planning and dietary recommendations, offering more accessible and cost-effective solutions. Nevertheless, further studies are needed to address its limitations and challenges.