With rising global temperatures posing a threat to agriculture, some wineries are turning to artificial intelligence to better prepare for unpredictable weather patterns.
“AI helps us be more resilient,” says Will Drayton, director of sustainability and science at Treasury Wine Estates. “So it’s not stopping climate change, it’s helping us be more resilient to climate change.”
Today, winemaker Beringer and Mattoa are using AI to help model “what-if” scenarios to more agilely plan for the big temperature swings that significantly impact the crop during the harvest season. AI can also help predict fermentation conditions and optimize irrigation methods.
Morning sunrise
Alcohol brands are still in the early stages of deploying AI, and beyond improvements in irrigation, a growing number of companies are using the technology to inform the chemistry that goes into making wine, beer or spirit, with the ultimate goal of creating new brands or flavours that are a hit with consumers.
The industry certainly needs a boost. Global demand for alcoholic beverages decreased In 2023, with little growth expected for this year and 2025, according to data and insights provider IWSR.
In the United States, younger adults are drink less alcohol The beer industry has become more influenced by previous generations, a trend that affects every industry differently. The beer industry has been bruised for years, as consumption has peaked. Lowest annual level since 1999Winemakers are struggling to attract millennial and Gen Z wine drinkers. Even the more resilient wine industry posted its first volume decline in 2023 in nearly 30 years, IWSR reports.
The Treasury says some of the more advanced technologies — including automation and robotics — it uses tend to favor larger brands, such as Beringer. But many of the AI investments are spread across the company’s portfolio in the U.S. and overseas markets including Australia and New Zealand.
The company uses artificial intelligence to achieve more precise irrigation. Sensors track water pressure and alert you when lines are inefficient, such as when wolves chew through drip lines. The data flows through AI models that make recommendations on when to fix the problems.
“With the 100-plus degree weather we have in California right now, paying attention to irrigation issues will clearly save yield and quality,” Drayton says.
Similarly, Sicilian winery Donnafugata is using the technology to help combat an invasion of European moths. Twice a day, a camera takes an image, which is processed by AI to identify the insects and predict their growth. This data helps devise a more aggressive defense strategy.
The winery also collects local climate data such as humidity, wind and barometric pressure, and combines it with weather forecasts up to 10 days out to help Donnafugata plan how to respond to damaging climate change. “The AI ultimately translates into adopting more sustainable practices,” says Antonio Rallo, Donnafugata’s co-owner and CEO.
Drinkers don’t always know what they like.
Katrina Axelsson, CEO and Founder of a Tech Startup TastyWineries often take unnecessary risks when launching new brands, he says. Historically, the industry looks at category data from analytics firm Nielsen to see what has sold well in the past, then develops a copycat brand in hopes of making a splash.
“I thought that was completely inadequate,” Axelson says.
Tastry uses AI to analyze wine blends and run simulations to determine how many different versions of that blend a winery can make, steering them toward the best possible chemistry that matches consumer preferences.
“Traditionally, it feels like consumers are being put into archetypes, with descriptive data that doesn’t necessarily relate to whether they like this wine or not,” Axelson says. “We see you as unique.”
Alcoholic beverage giant Diageo says one of the most advanced use cases for AI is predicting flavor preferences. In 2022, it acquired Vivienda and has since expanded intowhat is your whiskeyA platform that allows spirits drinkers to take a short quiz to discover which spirit best suits their taste.
The “What’s Your Whiskey” app is now available in 26 countries and 60 different languages, and similar AI tools have been unveiled for beer and cocktails. Tequila will be next.
“Being able to use this data with AI to more accurately understand where we should target our innovations, and even identify the right cocktails to serve in different locations… is a capability that will really stand us in good stead for the future,” says Mark Sandy, chief innovation officer at Digio.
It’s also led to some counterintuitive discoveries. AI data has shown that fruity whiskies are a hit with consumers, who tend to tell market researchers they prefer smoky flavors. Along those lines, Buchanan’s Pineapple Scotch was Diageo’s biggest North American launch this year.
Digio is also working with a startup called AI Palette, which feeds data from social media and food and drink trends into an AI system to predict the next hot flavor. This year, Digio is focusing on two flavors: umami and another called “flower harvest,” which includes jasmine, lavender and hibiscus.
“This is something we’re increasingly doing to try to predict where the consumer is going to go,” Sandy says, adding that Digio can then plan product innovation and promote cocktails that match the flavors drinkers are drawn to.
My beer analysis
Earlier this year, Belgian scientists conducted Published research The researchers analyzed 250 different types of beer and trained machine learning models to predict flavor and consumer preferences. For each beer, the scientists from research university KU Leuven measured 226 chemical properties, including alcohol and sugar content and flavors such as caramel, citrus, tropical fruit and banana.
They also analyzed over 180,000 consumer reviews from online rating platform RateBeer to train 10 different AI models.
“When it comes to machine learning, we have seen that you can predict flavour reasonably well,” says Michel Schruers, research associate and PhD candidate in microbiology at KU Leuven.
The AI models showed particular promise in predicting the flavor of non-alcoholic beers, and for alcoholic offerings, the models proved best at predicting acidity and alcohol content. Aroma proved more difficult, because the study was limited to Belgian breweries, which produce beers known for their yeast-based flavor.
In the US, the Bruery, Species X Beer Project, and Asbury Park Brewery have unveiled new beers created in collaboration with AI recipes with very funny labels, including “AI-IPA,” “AIPA,” and “Darw1n.”
Asbury Park Brewery founder Bob McLean says his New Jersey-based brewery was aiming to produce a new hazy IPA. As a starting point, it relied on a recipe created by Microsoft’s Copilot AI before making some adjustments to the alcohol content and grains used in the beer. The chatbot was aiming to make the beer much stronger than McLean wanted.
While AI-IPA is a strong seller at the brewery’s brewpub, McLean says AI-driven innovation won’t go that far, noting that “humans will continue to lead the way in creating new beers.”
Source link
