Nutritional analysis of AI-generated diet plans based on popular online diet trends

IF 4 2区农林科学 Q2 CHEMISTRY, APPLIED

Journal of Food Composition and Analysis Pub Date : 2025-05-27 DOI:10.1016/j.jfca.2025.107850

Hatice Merve Bayram , Sedat Arslan

{"title":"Nutritional analysis of AI-generated diet plans based on popular online diet trends","authors":"Hatice Merve Bayram , Sedat Arslan","doi":"10.1016/j.jfca.2025.107850","DOIUrl":null,"url":null,"abstract":"<div><div>This study aimed to evaluate the nutritional composition and consistency of 1500 kcal daily diet plans generated by four generative Artificial Intelligence (AI) tools (ChatGPT-4, ChatGPT-4o, Mistral, and Claude) based on five popular diet types identified via Google Trends (keto, paleo, Mediterranean, intermittent fasting, and raw). Each AI model was prompted with standardized requests, and the resulting menus were analyzed using Nutrition Information System (BeBIS) (version 9.0) to determine energy, macronutrient, and micronutrient content. Nutrient composition differences across AI tools were statistically assessed using SPSS 24.0 (ANOVA, p < 0.05). Results showed significant variations between AI outputs, with energy values ranging from 1357 kcal to 2273 kcal and protein intake varying by up to 65 g across models. Notable inconsistencies were also found in micronutrients such as calcium, iron, and vitamin D. AI models often failed to meet targeted caloric levels and showed inconsistent adherence to diet-specific nutrient profiles. These discrepancies suggest limitations not only in the AI tools’ capabilities but also in their interpretation of user prompts. The findings highlight the need for improved prompt design, database integration, and AI training for safe and reliable use in personalized nutrition.</div></div>","PeriodicalId":15867,"journal":{"name":"Journal of Food Composition and Analysis","volume":"145 ","pages":"Article 107850"},"PeriodicalIF":4.0000,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Food Composition and Analysis","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0889157525006659","RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, APPLIED","Score":null,"Total":0}

引用次数: 0

Abstract

This study aimed to evaluate the nutritional composition and consistency of 1500 kcal daily diet plans generated by four generative Artificial Intelligence (AI) tools (ChatGPT-4, ChatGPT-4o, Mistral, and Claude) based on five popular diet types identified via Google Trends (keto, paleo, Mediterranean, intermittent fasting, and raw). Each AI model was prompted with standardized requests, and the resulting menus were analyzed using Nutrition Information System (BeBIS) (version 9.0) to determine energy, macronutrient, and micronutrient content. Nutrient composition differences across AI tools were statistically assessed using SPSS 24.0 (ANOVA, p < 0.05). Results showed significant variations between AI outputs, with energy values ranging from 1357 kcal to 2273 kcal and protein intake varying by up to 65 g across models. Notable inconsistencies were also found in micronutrients such as calcium, iron, and vitamin D. AI models often failed to meet targeted caloric levels and showed inconsistent adherence to diet-specific nutrient profiles. These discrepancies suggest limitations not only in the AI tools’ capabilities but also in their interpretation of user prompts. The findings highlight the need for improved prompt design, database integration, and AI training for safe and reliable use in personalized nutrition.

查看原文本刊更多论文

基于在线流行饮食趋势的人工智能生成饮食计划的营养分析

本研究旨在评估由四种生成式人工智能（AI）工具（ChatGPT-4、chatgpt - 40、Mistral和Claude）生成的1500 千卡每日饮食计划的营养成分和一致性，该计划基于谷歌Trends确定的五种流行饮食类型（生酮饮食、古饮食、地中海饮食、间歇性禁食和生饮食）。每个AI模型都提示标准化的要求，并使用营养信息系统（BeBIS）（9.0版本）分析生成的菜单，以确定能量、宏量营养素和微量营养素的含量。人工智能工具间营养成分差异采用SPSS 24.0进行统计评估（方差分析，p <； 0.05）。结果显示，不同模型的人工智能输出之间存在显著差异，能量值从1357 kcal到2273 kcal不等，蛋白质摄入量的差异高达65 g。在钙、铁和维生素d等微量营养素中也发现了明显的不一致。人工智能模型经常无法达到目标热量水平，并且对饮食特定营养成分的坚持不一致。这些差异不仅表明人工智能工具的能力存在局限性，而且表明它们对用户提示的解释存在局限性。研究结果强调需要改进即时设计、数据库集成和人工智能培训，以便在个性化营养中安全可靠地使用。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Food Composition and Analysis 工程技术-食品科技

CiteScore

6.20

自引率

11.60%

发文量

601

审稿时长

53 days

期刊介绍： The Journal of Food Composition and Analysis publishes manuscripts on scientific aspects of data on the chemical composition of human foods, with particular emphasis on actual data on composition of foods; analytical methods; studies on the manipulation, storage, distribution and use of food composition data; and studies on the statistics, use and distribution of such data and data systems. The Journal''s basis is nutrient composition, with increasing emphasis on bioactive non-nutrient and anti-nutrient components. Papers must provide sufficient description of the food samples, analytical methods, quality control procedures and statistical treatments of the data to permit the end users of the food composition data to evaluate the appropriateness of such data in their projects. The Journal does not publish papers on: microbiological compounds; sensory quality; aromatics/volatiles in food and wine; essential oils; organoleptic characteristics of food; physical properties; or clinical papers and pharmacology-related papers.