- Written by Stav Dimitropoulos and Will Smale
- business reporter
Cala Maraden manages quite a few vineyards.
As director of viticulture at Foley Family Farms, a major U.S. wine company, she is responsible for more than 5,200 acres (2,000 hectares) of vineyards spread over 1,000 miles of California and Oregon.
Although Maladen can’t be everywhere at once, thanks to AI technology he can check the water needs of individual vineyards remotely from his laptop.
“You can go online and find out what type of irrigation is needed in Santa Barbara, 300 miles away,” said Maladen, who is based in Napa, northwest of San Francisco.
Determining the water needs of the vines was previously done through human judgment and calculations, but Foley now uses sensors from Tour Technologies, a California-based irrigation company.
Sensors, like mini weather stations, are installed throughout the vineyard. Measure humidity levels, temperature, wind speed, and other environmental variables.
All this data is fed into Tule’s AI software system, which is trained to calculate how much water evaporates from both the soil and the vines under different weather conditions. The AI then determines how much and when the vines need irrigation and notifies vineyard managers through app notifications.
“We always keep our feet on the ground,” Maladen added. “But data helps us make informed decisions based on science, not just emotions. Emotions are not a bad thing, but we prefer to use data.”
Foley Family Farms, along with its similarly named sister company Foley Family Wines, produces U.S. wine under 23 different brands. Ms Maladen says the AI irrigation sensors have “improved the quality and consistency of the grapes”.
Tule chief executive Tom Shapland says another benefit of the sensors is that they can help overcome labor shortages. “AI will monitor the vineyard 24/7.”
Tule is also developing an AI-powered app called Tule Vision. The app can determine how dry grapes are after a human user takes a few minutes of video of the vines. The AI was trained using hundreds of images of vines under different water demands.
Other providers of AI-powered vineyard monitoring equipment include US technology giant Cisco and smaller companies Ceres Imaging and Bloomfield AI.
Once the wine grapes are harvested, they are taken to the winery for fermentation. AI is starting to get involved here too.
Tastry, a California-based technology company, is developing an AI-powered app that helps winemakers create wines that millions of consumers will love. The software does this by analyzing the chemical composition of the wine and quickly comparing it to her database of taste preferences of 248 million wine lovers in the United States.
Tasting tests test the wine for everything from sugar content to acidity to tannins. The latter is a bitter and astringent compound derived from grape skins, seeds, and stems.
The idea is that it can help winemakers better blend a final wine from a number of base wines, varying the proportions of each until a popular flavor profile is achieved.
Alexandre Remy, managing partner and winemaker at California-based wine brand Atlas Wine Company, describes Tastry as a “GPS system.”
“If I want to make a red blend, I choose from 10 different vineyards,” says Remy. “This is where AI comes into its own. AI can help you decide what your goals are, whether you want to appeal to a younger demographic or a demographic that prefers your competitors’ wines.
“You input your parameters into the system and it gives you blending suggestions based on that.”
Monika Christmann is a professor of oenology, or the science of winemaking, at Geisenheim University in Germany. She says AI could be especially useful for wineries with high production volumes and a need for consistency from year to year.
However, she also cautions that the interactions between the various components of wine are difficult to predict and are not yet fully understood by AI.
For wine traditionalists, the use of AI in winemaking, whether in vineyards or wineries, is anathema. They argue that it destroys the craftsmanship that makes wine special.
Jonathan Kleeman, executive head sommelier at the Tom Sellers Storey Group, a British restaurant business, said: “Algorithms that can truly understand the complex interactions between climate, soil and grape varieties that give a wine its delicacy and character are “No,” he says.
“They also don’t understand the aesthetics of balance, structure and aromas that distinguish exceptional vintages. Telling a wine how mineral, acidic or fruity it is is meaningless to a machine. A true winemaker trusts his palate. To do. [computer] processor. ”
Tom Ashworth, CEO of wine retailer Yap Brothers, said he wouldn’t use AI if he was making wine.
“I think the extent to which winemakers can delegate decision-making to AI – replacing hundreds of years of experience – will depend on the risk appetite of the domaine and the accuracy of the AI itself.
“At this point, we have no intention of handing over decision-making for key processes in brewing to AI, any more than we would let it perform customer service without oversight.”
Back in California, Maladen counters that AI is meant to help human winemakers, not replace them. “We are boots on the ground. We are always in the vineyard ourselves,” she says.
“What exactly is AI?” [in wine] We’re making more informed decisions based on science. ”