“Using the different sets of data, the team constructed models based on machine learning – a form of AI – to predict how a beer would taste, and its appreciation, based on its composition.
They then used the results to enhance an existing commercial beer, essentially spiking it with substances flagged by the models as being important predictors of overall appreciation – such as lactic acid and glycerol.
The results from the tasting panel revealed the additions improved ratings for both alcoholic and non-alcoholic beers across metrics including sweetness, body, and overall appreciation.”
From The Guardian.