
Researchers at Stanford University in California have developed an artificial intelligence tool that designs optimal burger recipes tailored to users' preferences on taste, sustainability and nutritional content, according to a study published in the scientific journal npj Science of Food.
The AI-generated recipes also passed a real-world taste test with restaurant diners, the research team reports.
"AI did not just generate plausible burger recipes - it created burgers that real people enjoy," said Ellen Kuhl, a bioengineering professor at Stanford. "That may sound simple, but it means the model learned what makes food appealing to the human palate." That was by no means trivial, she added.
The modern hamburger emerged in the late 19th century as a simple combination of minced meat and bread. Kuhl wrote in the journal Computer Methods in Applied Mechanics and Engineering that although around 50 billion hamburgers are now consumed in the US alone each year, the possible combinations of burger ingredients remain largely unexplored.
It is no surprise: the team estimates the number of possible burger recipes worldwide at 10 septillion - that is a 1 followed by 43 zeros. To bring some order to these seemingly endless combinations, Kuhl and two colleagues designed the publicly accessible online tool BurgerAI, trained on 2,216 recipes.
The software calculates not only the taste of a burger using mathematical principles, but also its texture, nutritional content and sustainability - that is, its environmental impact - for different age groups, genders and consumer lifestyles.
"For centuries, food design has been a matter of intuition, experience and trial and error," said Kuhl. "We are beginning to show that AI can transform food design into a quantitative science with applications in other important fields."
The ultimate test was a blind trial involving 101 volunteers at a restaurant in San Francisco. They were served five AI-designed burgers, professionally prepared, alongside a Big Mac for comparison.
The mushroom burger outperformed the conventional product on sustainability by more than tenfold, while the bean burger surpassed it on nutritional value by almost double, the trio of researchers writes.
The AI burgers also held their own on taste in direct comparison: in terms of flavour, texture and overall verdict, they were at least on a par with the established burger classic, according to ratings on a seven-point scale.
"We expected some trade-off between sustainability and consumer acceptance," said lead author Vahidullah Tac. "But we found a burger with dramatically lower environmental impact could still compete with one of the world’s most successful burgers."
Food choices are among the most consequential everyday decisions people make, he stressed: "With one arrow, you can hit two targets – planetary health and personal health. It’s a great and impactful research area."

