ChatGPT Provides Accurate Nutritional Information, Research Shows

This article has been reviewed in accordance with Science X’s editorial process and policies. The editors have emphasized the following attributes while ensuring the credibility of the content:

verified facts

peer-reviewed publication

renowned press agency

reread

ChatGPT can be a useful and convenient tool for people who want to know the nutritional information of specific foods, according to a research letter published online Dec. 27 in Open JAMA Network.

Yen Nhi Hoang, from Taipei Medical University in Taiwan, and colleagues studied the reliability of artificial intelligence (AI) in providing the energy and macronutrient contents of 222 food products in English and Chinese. The reliability of ChatGPT-3.5 (chatbot 1) and ChatGPT-4 (chatbot 2) was compared in providing information on calorie and macronutrient (carbohydrate, fat, and protein) content.

The researchers observed no significant difference between the nutritionists’ and AI’s estimates of energy, carbohydrate, and fat content, but there was a significant difference in the protein estimate. Both versions of ChatGPT provided accurate energy contents for approximately 35–48% of the 222 food items within a 10% range, with coefficient variation <10%. Chatbot 2 performed better than chatbot 1, but overestimated protein.

“Although AI chatbots are designed to be probabilistic, the results of this cross-sectional study suggest that AI can be a useful and practical tool for people who want to know the energy and macronutrient information of their foods,” write the authors. “Currently, the ability of AI chatbots to provide personalized dietary advice, such as specific nutritional guidelines and exact portion sizes, is limited.”

More information:
Yen Nhi Hoang et al, Consistency and accuracy of artificial intelligence to provide nutritional information, Open JAMA Network (2023). DOI: 10.1001/jamanetworkopen.2023.50367

Journal information:
Open JAMA Network

Related Articles

wordpress reviews