The role of large language models in personalized obesity management: opportunities and challenges

Abstract—Background: Obesity is a worldwide public health issue that needs to be treated individually, with the use of multidisciplinary therapy. Large Language Models (LLMs), like ChatGPT, provide large-scale AI-supported solutions for nutritional planning, physical activity prescription, and behavioral treatment. Their safety, accuracy, and effectiveness, nonetheless, have to be tested. Objective: This systematic review introduces the potential of LLMs in the treatment of obesity and their advantages, disadvantages, and ethics. Methods: The PubMed, Scopus, and IEEE Xplore databases were searched from 2023 to 2025 for peer-reviewed journal articles on the use of LLMs for weight management following PRISMA standards. Results: The promise of LLMs was discovered to discover individualized diet and exercise regimens, promote user engagement, and provide pharmacotherapeutic management of obesity. But there are issues of AI-generated advice reliability, user input quality dependence, absence of human control, and ethical issues like data privacy and accountability. Conclusion: Though LLMs enhance availability for obesity control, their potential risks need stringent control. Hybrid human-AI offers a solution that can maximize benefits while securing safety and efficacy. Future studies need to enhance transparency in AI, formulate regulatory systems, and promote cultural flexibility for equitable health utilization.