Can the AI tools ChatGPT and Bard generate energy, macro- and micro-nutrient sufficient meal plans for different dietary patterns?

Abstract

Artificial intelligence (AI) chat bots based on large language models have recently emerged as an alternative to traditional online searches and are also entering the nutrition space. In this study we wanted to investigate whether the AI chat bots ChatGPT and Bard (now Gemini) can create meal plans that meet the dietary reference intake (DRI) for different dietary patterns. We further hypothesized that nutritional adequacy could be improved by modifying the prompts used. Meal plans were generated by three accounts for different dietary patterns (omnivorous, vegetarian and vegan), using two distinct prompts resulting in 108 meal plans total. The nutrient content of the plans was subsequently analyzed and compared to the DRIs. On average, the meal plans contained less energy and carbohydrates but mostly exceeded the DRI for protein. Vitamin D and fluoride fell below the DRI for all plans while only the vegan plans contained insufficient vitamin B12. ChatGPT suggested using vitamin B12 supplements in 5 out of 18 instances, while Bard never recommend supplements. There were no significant differences between the prompts or the two models. While the meal plans generated by ChatGPT and Bard met most DRIs, there were some exceptions, particularly for vegan diets. These models may be a useful tool for individuals looking for general dietary inspiration, but they should not be relied upon to create nutritionally adequate meal plans, especially for individuals with restrictive dietary needs.