Stanford researcher Ellen Kuhl estimates there are about 1043 possible burger recipes in the world. And with BurgerAI, a new tool developed in its lab, AI can now design the best for you based on your age, taste, nutritional needs and even your sustainability goal.
But BurgerAI’s ability to recommend a great-tasting, nutritionally complex, sustainably produced burger is only part of the story. More broadly, this innovation heralds a change for AI itself: moving AI from prediction to design.
“Most AI systems are trained to predict what’s already there. We wanted AI to do that invent what should be next,” explained Kuhl, a professor of mechanical engineering in the School of Engineering who now directs Stanford Bio-X, an interdisciplinary life sciences institute that brings together researchers from medicine, engineering and the natural sciences. He asks, “Which burger best meets these important and complex goals?”
Food in focus
Food is the next big thing in the life sciences, Kuhl said, a focus that combines elements of human experience and culture, health and nutrition, and environmental impact, which are topics that inspire multidisciplinary researchers from the schools of medicine, engineering, sustainability, humanities and sustainability and beyond.
“Food choices are some of the most important decisions people make every day,” said Vahidullah Tac, a Schmidt Science postdoctoral fellow in Kuhl’s lab. “Food was an easy motivator. With one arrow, you can hit two targets – the health of the planet and personal health. It’s a great and exciting area of research.”
Therefore, food proved to be an ideal test bed for Bio-X. Kuhl’s team has just published two papers on BurgerAI, of which Tac is the first author. The first article introduces BurgerAI. The second paper reveals that the same mathematical principles that drive BurgerAI also underpin diffusion-based genetic AI more broadly and make connections to technical fields such as materials design, physics and engineering.
For centuries, food design has been a matter of intuition, experience and trial and error. We are beginning to show that AI can turn food design into a quantitative science with applications in other important areas.”
Ellen Kuhl, Professor of Mechanical Engineering, School of Engineering, Stanford University
Taste tested
Using 2,216 burger recipes from Food.com as a data source, BurgerAI learns patterns in ingredient combinations and quantities and then creates new burger recipes from scratch. AI then matches these characterizations with human taste and texture profiles. The results are completely new recipes optimized for taste, sustainability and nutrition, and personalized by gender, age and physical activity.
The ultimate test was not computing but cooking. The researchers served five professionally prepared, AI-designed hamburgers to more than 100 customers in a blind taste test at a San Francisco restaurant. In a side-by-side comparison with a popular fast-food burger, the two variants of BurgerAI’s Delicious Burger scored the same or better for overall likability, taste and texture. Its Mushroom Burger reduced its environmental impact by more than an order of magnitude, and its Bean Burger achieved about twice the nutritional score of the fast-food burger.
“Artificial intelligence didn’t just create plausible burger recipes — it created burgers that real people enjoy,” Kuhl said. “This may sound simple, but it means the model learned what makes food appealing to the human palate and was able to navigate a design space of nearly infinite possible burger combinations to find real-world solutions.”
Beyond the burgers
Tac was really surprised by how well the sustainable burgers performed. “We expected some trade-off between sustainability and consumer acceptance,” he said. “But we found that a burger with a dramatically lower environmental impact could still compete with one of the most successful hamburgers in the world.”
For Tac and Kuhl, BurgerAI isn’t really about hamburgers. It’s a proof of concept for the broader design possibilities of AI. The same genetic design framework could have implications for other downstream areas – pharmaceuticals, materials, biomolecules, and other complex systems with vast design spaces. As with food, which requires a balance of taste, nutrition, cost and sustainability, many of society’s biggest challenges must balance competing goals. If AI can help navigate trade-offs in recipe design, Kuhl said, it could also help discover new drugs, make advanced materials and create more sustainable products.
“The burger is just the beginning,” assured Kuhl. “We see food as a model system for a much bigger vision: artificial intelligence as a partner in scientific and engineering discovery.”
