Stanford University researchers have made a breakthrough by training an artificial intelligence (AI) system to create burger recipes that diners rated on par with, or even surpassing, those from a renowned global fast-food chain. A study published in npj Science of Food reveals that their AI tool, BurgerAI, was trained using 2,216 burger recipes from Food.com.
AI-Powered Taste Innovation
BurgerAI generated several burger recipes that either matched or outperformed a re-engineered version of McDonald’s Big Mac in blind taste tests. Remarkably, the Big Mac itself was not included in the initial training data. The AI achieved similar flavors by learning from collective taste preferences.
The AI tool not only focuses on taste but also allows for customization. It can create burgers that are healthier, environmentally friendlier, and personalized based on demographics and lifestyle, as detailed in a Stanford Report news release.
Broader Impacts and Discoveries
Researcher Vahidullah Tac highlighted the dual benefits of such innovations, emphasizing both health and ecological advantages. The project aims to use AI as a partner in scientific and engineering innovations. Stanford Professor Ellen Kuhl explained that unlike conventional AI, which predicts existing scenarios, their AI system seeks to invent optimally satisfying outcomes.
The AI-generated burgers underwent evaluation by over 100 diners during blind taste tests in a San Francisco restaurant. The results showed that the burgers, particularly the mushroom burger, scored highly in taste and offered a lower environmental impact.
Future Applications and Insights
Kuhl believes the success of BurgerAI marks the beginning of AI’s transformative role in food design, making it a precise science. According to her, AI’s potential extends beyond food into various fields such as pharmaceuticals and product design.