AI's Nose for Wine: Machine Learning Sniffs Out Fraudulent Bordeaux

Scientists have developed a machine learning algorithm that can analyze the chemical compounds in wine to detect fraud and trace the wine back to its estate of origin. Researchers trained an artificial intelligence on data from 80 wines from 7 Bordeaux estates over 12 years. The AI detects patterns across many compounds, not just a single compound, to determine a wine's unique signature.

The AI was able to cluster wines by estate and match them to their geographic origin with 99% accuracy. The research could help combat the $3 billion wine fraud industry in Europe, where criminal networks pass off cheap wine as expensive vintages.

"There's a lot of wine fraud around, with people making up some crap in their garage, printing off labels and selling it for thousands of dollars," said Prof. Alexandre Pouget at the University of Geneva, who led the research. "We show for the first time that we have enough sensitivity with our chemical techniques to tell the difference."

The machine learning algorithm draws on all the chemicals detected in a wine to determine its signature. The program displays the results on a two-dimensional grid, where wines with similar signatures group together.

"The first thing we saw that jumped straight out at us is there are clusters that correspond to a specific châteaux," Pouget said. "That told us right away that there is a chemical signature specific to each châteaux, independent of vintage. It's the overall pattern of concentrations of many, many molecules that distinguishes a châteaux. Each is a symphony; there isn't a single note that distinguishes them, it's the whole melody."

The research suggests machine learning could aid fraud investigations by confirming whether a wine's chemical signature matches what's on its label. While the algorithm traced wines to the correct estates with 99% accuracy, it struggled to distinguish vintages, reaching only 50% accuracy.

The approach could also help monitor quality and blend wines to optimize taste. "Wine blending is the key step in making great Bordeaux and Champagne," Pouget said. "So far this is done by a few winemakers who are paid a fortune for their skills. Having tools like this would make it a lot cheaper to make great blends, which would benefit everybody."

Protect your palate and pocketbook - demand wine fraud detection to ensure you get what you pay for. Contact your political representatives and ask them to support regulations and funding to combat wine fraud. Together, we can stamp out counterfeit wines.