URBANA, IL (Chambana Today) – A new study from the University of Illinois Urbana-Champaign explores how AI, specifically ChatGPT, can be used in the sensory evaluation of foods, with a focus on brownies. The research highlights how AI could streamline the development of new food products by offering insights into flavor, texture, and overall appeal without the need for expensive and time-consuming human taste tests. Torrico believes AI could help food scientists by acting as a preliminary screening tool, identifying recipes worth further human testing.
In the study, Damir Torrico, assistant professor in the Department of Food Science and Human Nutrition, evaluated 15 different brownie recipes, ranging from traditional to those with unusual ingredients like mealworm powder and fish oil. Torrico fed these recipes into ChatGPT, which was tasked with describing sensory characteristics such as taste and texture. Surprisingly, even for the odd ingredient combinations, ChatGPT’s responses were overwhelmingly positive, reflecting a psychological phenomenon known as hedonic asymmetry, where things perceived as beneficial are described more positively.
Torrico believes AI could help food scientists by acting as a preliminary screening tool, identifying recipes worth further human testing. Though challenges like hedonic asymmetry remain, Torrico sees the potential for AI to save both time and money in product development. While ChatGPT may not be ready to taste-test your brownies, Torrico is optimistic about its future in food science.
The study, “The Potential Use of ChatGPT as a Sensory Evaluator of Chocolate Brownies: A Brief Case Study,” is published in Foods.