How farmers are using AI on vineyards to make wine

When artificial intelligence-backed tractors became available to vineyards, Tom Gamble wanted to be an early adopter. He knew there would be a learning curve, but Gamble decided the technology was worth figuring out.

The third-generation farmer bought one autonomous tractor. He plans on deploying its self-driving feature this spring and is currently using the tractor’s AI sensor to map his Napa Valley vineyard. As it learns each row, the tractor will know where to go once it is used autonomously. The AI within the machine will then process the data it collects and help Gamble make better-informed decisions about his crops—what he calls “precision farming.”

“It’s not going to completely replace the human element of putting your boot into the vineyard, and that’s one of my favorite things to do,” he said. “But it’s going to be able to allow you to work more smartly, more intelligently and in the end, make better decisions under less fatigue.”

Gamble said he anticipates using the tech as much as possible because of “economic, air quality, and regulatory imperatives.” Autonomous tractors, he said, could help lower his fuel use and cut back on pollution.

As AI continues to grow, experts say that the wine industry is proof that businesses can integrate the technology efficiently to supplement labor without displacing a workforce. New agricultural tech like AI can help farmers to cut back on waste, and to run more efficient and sustainable vineyards by monitoring water use and helping determine when and where to use products like fertilizers or pest control. AI-backed tractors and irrigation systems, farmers say, can minimize water use by analyzing soil or vines, while also helping farmers to manage acres of vineyards by providing more accurate data on the health of a crop or what a season’s yield will be.

Other facets of the wine industry have also started adopting the tech, from using generative AI to create custom wine labels to turning to ChatGPT to develop, label, and price an entire bottle.

“I don’t see anybody losing their job, because I think that a tractor operator’s skills are going to increase and as a result, and maybe they’re overseeing a small fleet of these machines that are out there, and they’ll be compensated as a result of their increased skill level,” he said.

Farmers, Gamble said, are always evolving. There were fears when the tractor replaced horses and mules pulling plows, but that technology “proved itself” just like AI farming tech will, he said, adding that adopting any new tech always takes time.

Companies like John Deere have started using the AI that wine farmers are beginning to adopt. The agricultural giant uses “Smart Apply” technology on tractors, for example, helping growers apply material for crop retention by using sensors and algorithms to sense foliage on grape canopies, said Sean Sundberg, business integration manager at John Deere.

The tractors that use that tech then only spray “where there are grapes or leaves or whatnot so that it doesn’t spray material unnecessarily,” he said. Last year, the company announced a project with Sonoma County Winegrowers to use tech to help wine grape growers maximize their yield.

Tyler Klick, partner at Redwood Empire Vineyard Management, said his company has started automating irrigation valves at the vineyards it helps manage. The valves send an alert in the event of a leak and will automatically shut off if they notice an “excessive” water flow rate.

“That valve is actually starting to learn typical water use,” Klick said. “It’ll learn how much water is used before the production starts to fall off.”
Klick said each valve costs roughly $600, plus $150 per acre each year to subscribe to the service.

“Our job, viticulture, is to adjust our operations to the climatic conditions we’re dealt,” Klick said. “I can see AI helping us with finite conditions.”

Angelo A. Camillo, a professor of wine business at Sonoma State University, said that despite excitement over AI in the wine industry, some smaller vineyards are more skeptical about their ability to use the technology. Small, family-owned operations, which Camillo said account for about 80% of the wine business in America, are slowly disappearing—many don’t have the money to invest in AI, he said. A robotic arm that helps put together pallets of wine, for example, can cost as much as $150,000, he said.

“For small wineries, there’s a question mark, which is the investment. Then there’s the education. Who’s going to work with all of these AI applications? Where is the training?” he said.

There are also potential challenges with scalability, Camillo added. Drones, for example, could be useful for smaller vineyards that could use AI to target specific crops that have a bug problem, he said—it would be much harder to operate 100 drones in a 1,000-acre vineyard while also employing the IT workers who understand the tech.

“I don’t think a person can manage 40 drones as a swarm of drones,” he said. “So there’s a constraint for the operators to adopt certain things.”

However, AI is particularly good at tracking a crop’s health—including how the plant itself is doing and whether it’s growing enough leaves—while also monitoring grapes to aid in yield projections, said Mason Earles, an assistant professor who leads the Plant AI and Biophysics Lab at UC Davis.

Diseases or viruses can sneak up and destroy entire vineyards, Earles said, calling it an “elephant in the room” across the wine industry. The process of replanting a vineyard and getting it to produce well takes at least five years, he said. AI can help growers determine which virus is affecting their plants, he said, and whether they should rip out some crops immediately to avoid losing their entire vineyard.

Earles, who is also cofounder of the AI-powered farm management platform Scout, said his company uses AI to process thousands of images in hours and extract data quickly—something that would be difficult by hand in large vineyards that span hundreds of acres. Scout’s AI platform then counts and measures the number of grape clusters as early as when a plant is beginning to flower in order to forecast what a yield will be.

The sooner vintners know how much yield to expect, the better they can “dial in” their wine making process, he added.

“Predicting what yields you’re going to have at the end of the season, no one is that good at it right now,” he said. “But it’s really important because it determines how much labor contract you’re going to need and the supplies you’ll need for making wine.”

Earles doesn’t think the budding use of AI in vineyards is “freaking farmers out.” Rather, he anticipates that AI will be used more frequently to help with difficult field labor and to discern problems in vineyards that farmers need help with.
“They’ve seen people trying to sell them tech for decades. It’s hard to farm; it’s unpredictable compared to most other jobs,” he said. “The walking and counting, I think people would have said a long time ago, ‘I would happily let a machine take over.'”

—Sarah Parvini, AP technology writer

https://www.fastcompany.com/91294228/farmers-using-ai-vineyards-to-make-wine?partner=rss&utm_source=rss&utm_medium=feed&utm_campaign=rss+fastcompany&utm_content=rss