21 March 2024 | Dong Wook Kim, Ji Seok Park, Kavita Sharma, Amanda Velazquez, Lu Li, John W. Ostrominski, Tram Tran, Robert H. Seitter Perez and Jeong-Hun Shin
This study evaluates the potential of AI-generated personalized weight-loss diet plans for clinical applications. Utilizing ChatGPT (4.0), the researchers created diet plans and compared them with two control plans from tertiary medical centers. Experts in obesity medicine and clinical nutrition provided feedback on the effectiveness, balancedness, comprehensiveness, flexibility, and applicability of the plans. The results showed no significant differences between the AI-generated and control plans in any evaluation category. Only five out of 14 experts correctly identified the AI-generated plan. The AI-generated personalized diet plan received scores above neutral in all evaluation variables, but limitations were highlighted, including conflicting dietary considerations, lack of affordability, and insufficient specificity in recommendations. The study concludes that while AI-generated diet plans are promising, further refinement and prospective studies are needed to enhance their quality and applicability in real-world clinical settings.This study evaluates the potential of AI-generated personalized weight-loss diet plans for clinical applications. Utilizing ChatGPT (4.0), the researchers created diet plans and compared them with two control plans from tertiary medical centers. Experts in obesity medicine and clinical nutrition provided feedback on the effectiveness, balancedness, comprehensiveness, flexibility, and applicability of the plans. The results showed no significant differences between the AI-generated and control plans in any evaluation category. Only five out of 14 experts correctly identified the AI-generated plan. The AI-generated personalized diet plan received scores above neutral in all evaluation variables, but limitations were highlighted, including conflicting dietary considerations, lack of affordability, and insufficient specificity in recommendations. The study concludes that while AI-generated diet plans are promising, further refinement and prospective studies are needed to enhance their quality and applicability in real-world clinical settings.