To those of us who taste wine for a living, it was both fascinating and alarming to read last week’s news that researchers in Switzerland have trained an artificial intelligence program to identify Bordeaux wines from specific estates and vintages with 100% certainty.
In the study, published in the journal Communications Chemistry, scientists at the University of Geneva used machine learning to analyze the chemical composition of 80 red wines from 12 vintages between 1990 and 2007. “We were interested in finding out whether there is a chemical signature that is specific to each of those chateaux that’s independent of vintage,” says lead researcher Alexandre Pouget in the New Scientist. They asked a question at the core of wine knowledge: Does an individual chateau’s wine have a similar chemical profile—and therefore taste—year after year?
Pouget and his colleagues vaporized the 80 wines and used chromatography to separate and catalog the wines’ chemical compounds. Each wine’s readout, or chromatogram, had 30,000 points representing different chemical compounds. The researcher then used 73 of the wine’s chromatograms to train a machine learning algorithm, along with data on the estate and vintage. Finally, they tested the A.I. algorithm on the remaining seven wines—50 times each in various orders. The A.I. blind tasting algorithm identified the wines 100% of the time. The algorithm also was able to group the wines based on whether they were from Bordeaux’s Left or Right Bank.
You May Also Like: Producers Feared the 2022 Bordeaux Vintage Would Be a Dud. It’s Exceptional.
“The fact that we could perfectly identify estates, independently of vintages, suggests that the estates we have analyzed here have distinct identities,” reads the study. “While wine experts believe that some Bordeaux estates have indeed distinct profiles, this is, to our knowledge the first time that this is demonstrated with a purely chemical analysis of Bordeaux wines.”
There is quite a bit to unpack here. One positive is that the study offers solid scientific proof that terroir exists. It’s something that many aficionados already know, that well-made wines exude a sense of place and that the surrounding environment deeply influences the end result. Maybe we can finally move on from stupid debates over whether terroir “matters” as well as the legions of terroir deniers who still populate the wine industry.
Beyond the terroir finding, however, these experiments might strike some fear into the people who base their personalities on blind tasting. In the concluding discussion section of the study, the scientists even make a cheeky suggestion: “It would be interesting to compare the performance of our model to the one of expert human tasters on blind tasting of the 80 wines we have analyzed. Whether expert wine tasters would be able to match our model’s performance (100% correct) on these seven estates is not known.” Um, as a human taster I hate to say it, but I think we know.
You May Also Like: A Ph.D. in Sensory Science Explains How to Improve Your Tasting Abilities
When we thought about A.I. in the past, we always assumed it would be used in the future to do the boring, droning, dangerous work that humans didn’t want to do. For many of us, it’s been unnerving, now that we’ve arrived in that future, to see that A.I. is in fact being used to do some things we thought were innately human—to write, create art, make music… and perhaps now “taste” wine.
But from another perspective, an A.I. wine critic might be a blessing in disguise. Wine has gotten so bogged down in our human attempts at being rational, logical and quasi-scientific. The current model of wine professionals, which is based on blind tasting skills, or the critic who tastes vast amounts of wine in one sitting and offers numerical scores both stem from the impulse of making our understanding of wine more “objective.” But wine isn’t any more objective than other human creations.
Maybe A.I. can free up the sommelier or critic to focus more on the emotional, romantic side of wine. I’ve advocated before for a new kind of wine criticism, something like the romantic criticism that art critic Morgan Meis has been espousing. A decade ago, Meis started advocating for a criticism that surrenders “all great claims to authority” and “refuses to judge” and whose primary virtue is “its inherent generosity.” Romantic criticism is not based on an infallible, all-knowing expert or unquestioned arbiter of taste.
“In this theory of criticism we don’t need the critic to tell us what is good or bad, to tell us what to like and dislike,” Meis writes.“We need the critic, instead, to help us experience. We need the critic in the way that we need a friend or a lover. We need the critic as a companion on a journey that is a love affair with the things of the world.”
By leaving the blind tasting and quantifications to the robots, perhaps we can reclaim what’s human about wine.
The post A.I. Can Identify Wine 100% of the Time. What Now? appeared first on Wine Enthusiast.
