We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.


AI Can Now Predict the Taste of Beer

Different beers.
Credit: Jon Parry/Unsplash
Listen with
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 1 minute

Artificial intelligence (AI) may just have outmoded another profession: beer sommeliers.

Researchers from the Leuven Institute for Beer Research in Belgium have developed AI models that can predict how a beer will taste based on its chemical makeup. The technology can even recommend how an ale could be improved.

Their findings were published in Nature Communications.

Open the fridge-cooled Coors, HAL

To bring some scientific rigor to the world of beer tasting, the Belgian researchers first analyzed the chemical makeup of 250 Belgian beers, which comprised of 22 different styles including lagers, ales and non-alcoholic beers.

“The flavor of beer is a complex mix of aroma compounds,” said Dr. Miguel Roncoroni, who led the chemical analyses and tasting panel. “It is impossible to predict how good a beer is by just measuring one or a few compounds. We really need the power of computers.”

Gas chromatography-mass spectrometry (GC-MS) was used to measure volatile compounds like terpenoids and esters; discrete photometric and enzymatic analysis was used to measure levels of acetic acid, ammonia, color, sugars, pH and sulfites. Sixteen trained volunteers also tasted the beers and recorded their sensory attributes like sweetness and acidity.

The AI was then fed 181,025 reviews from the RateBeer website. An inbuilt dictionary helped the computer model group similar adjectives, such as “floral” and “flower”.

“We began the project with less than 100 beers, and quickly realized this would not be enough to capture Belgium’s incredible beer diversity, so we ended up analyzing 250 beers,” said Roncoroni. “It was a truly Herculean effort.”

After five years of collecting and inputting data, the team was finally satisfied that their AI could reliably predict the key aromas and the final appreciation score of a beer. The model could even recommend additional ingredients for beers to improve their quality. When these alternative beers were produced, sure enough, the same panel of testers concurred that the flavors were indeed superior.

The team now intends to use their AI to help produce better-tasting alcohol-free beers.

“Our biggest goal now is to make better alcohol-free beer. Using our model, we have already succeeded in creating a cocktail of natural aroma compounds that mimic the taste and smell of alcohol without the risk of a hangover,” said Kevin Verstrepen, a professor in genetics and genomics at Leuven University.

Reference: Schreurs M, Piampongsant S, Roncoroni M et al. Predicting and improving complex beer flavor through machine learning. Nat. Commun. 2024. doi: 10.1038/s41467-024-46346-0

This article is a rework of a press release issued by the Vlaams Instituut voor Biotechnologie. Material has been edited for length and content.