A Japanese chain of sushi restaurants is using an AI-powered app to assess the quality of tuna — a key step in the preparation of sushi that traditionally requires years of training from experienced human buyers. But can it really replace a human’s fish sense?
The app, named Tuna Scope, was developed by Japanese advertising firm Dentsu Inc. It uses machine learning algorithms trained on thousands of images of the cross-sections of tuna tails, a cut of the meat that can reveal much about a fish’s constitution.
The app was trained on thousands of images of tuna tails
From a single picture, the app grades the tuna on a five-point scale based on visual characteristics like the sheen of the flesh and the layering of fat. For an experienced fish grader, these attributes speak volumes about the sort of life the fish led, what it ate, and how active it was — thus, the resulting flavor. Dentsu claims that its AI has captured the “unexplainable nuances of the tuna examination craft,” and in tests comparing the app with human buyers, the app issued the same grade more than four times out of five.
But sushi experts and fishmongers are a little more cautious about Tuna Scope’s ability to replace fish graders, especially those buying meat for high-end sushi and sashimi.
Dentsu’s app uses machine learning algorithms to assess the tuna from just a single picture. Image: Dentsu
Keiko Yamamoto, a chef and sushi instructor based in London, told The Verge that it’s certainly possible to grade tuna based on visuals alone. Although we often judge the quality of produce based on touch, Yamamoto says with tuna, appearance is everything. “I’ve had to cut fresh tuna every two weeks, so I know what’s good, what’s not good,” she says.
It’s hard to describe high-quality tuna, but chefs know it when they see it
Yamamoto says the exact qualities buyers are looking for can be hard to capture in words but are unmistakable to the trained eye. The highest-quality tuna has an intense bright red color and a certain degree of translucency, as if the flesh is almost glowing. “It looks bouncy, or soft, maybe, to your eye,” she says. “Good quality tuna is silkier and shiny.”
It seems possible to use AI to make basic assessments of the quality, says Yamamoto. She adds that she’s also not surprised that Japan is pursuing this tech, considering its aging population means traditional skills are not always passed down to younger generations.
Right now, according to The Asahi Shimbun, it seems Tuna Scope is only being used to grade fish for restaurant chain Kura Sushi, which offers cheap sushi and uses other cost-saving devices like robotic dishwashers. Kura Sushi reportedly purchases 70 percent of its fish for sushi overseas and is wary about its buyers traveling during the current pandemic. The app means local agents can make on-the-spot assessments instead.
Chefs and fishmongers judge the quality of tuna based on the sheen and color of the flesh. Photo by Tomohiro Ohsumi / Getty Images
But while this sort of automation might work for a large chain like Kura Sushi, it won’t meet the demands of high-end chefs and sushi aficionados, according to Richard Cann, a sales manager at T&S Enterprises, a wholesale fishmonger that supplies many of London’s top Japanese restaurants, like Nobu in Mayfair and Zuma in Knightsbridge.
“We always have and always will do it by eye,” Cann tells The Verge. “I don’t think there’s a need to grade tuna with an app.”
“I don’t think there’s a need to grade tuna with an app.”
This is partly to do with differences in the procurement process between chains and high-end restaurants. In Tuna Scope’s marketing material, buyers use the app to judge the quality of frozen tuna by snapping pictures of the tail section. But Cann says outfits like T&S Enterprises buy the tuna whole and unfrozen and divide it themselves into specific cuts.
In busy periods, Cann says his team receives two shipments a week of around fresh four tuna apiece, each of which can weigh upward of 500 pounds (226 kg) and has to be butchered by hand. Assessing the quality of the fish is not something that happens once, he says; it’s an ongoing process. “The guys who cut up our bluefin here, they’ve been doing it for 10, 15 years,” he says. “It’s a knack you pick up, you just know what’s good and know what’s bad.”
Cann says T&S has relationships with chefs around the city who trust its workers and, by extension, the quality of its fish. Trying to automate even part of the buying experience would break that chain of trust, he says. Because although the trade might be selling fish, “we’re a people business in everything we do.”
“We’d never use an app because we quite like human beings,” says Cann. “It’s good to have them around.”