While terrifyingly sophisticated robots a la Westworld are still a long way off, artificial intelligence technology has found plenty of applications in the food and restaurant sphere — from making restaurant recommendations and reducing food waste to recognizing customer’s faces and ensuring pizza toppings are evenly distributed.
And one startup believes that A.I. holds the key to the future of flavor: As the New Food Economy reports, NYC-based Analytic Flavor Systems envisions a future in which food is hyper-customized for people’s individual tastes — “a day when we’ll each have a Doritos of our own.”
Mass-market snacks and drinks are designed to appeal to as wide an audience as possible, resulting in what CEO Jason Cohen believes are “endless shelves of products that most people like, but few people really love.” His company has created an AI platform called Gastrograph — a free version can be downloaded via the iTunes App Store — that collects highly specific data from individual users, with the goal of “giving food and beverage companies the information they need to develop products optimized for more and more specific sensibilities.”
Here’s how Gastrograph works, according to TNFE:
The app’s central feature is a wheel with 24 spokes, where each sliver represents a discrete category of sensory experience—such as “meaty,” “bitter,” or “mouthfeel.” Tasters map the contours of flavor perception by tracing the spokes corresponding to the qualities they detect, designating the intensity of each on a scale from one to five. A submenu allows for a more granular record of experience: specifying that “meaty” quality, for instance, as beefy, sausage-like, or more exotic options (moose, kangaroo). Tasters are then prompted to give the product a preference rating, on a scale from one to seven.
In addition to information on flavor preferences, the app also collects objective data like the person’s socioeconomic status, whether or not they smoke, and even information about their environment like temperature and noise levels, all of which are thought to affect taste.
Cohen points to how big food and drink manufacturers currently conduct consumer taste tests as part of their product development process, noting that the consumer groups are often homogenous — he specifically cites a Molson Miller Coors tasting panel comprised entirely of white people — which makes them ill-suited to develop products for a large and diverse demographic.
Gastrograph has its critics, however, who argue that collecting data indiscriminately via an app doesn’t necessarily mean that data will be more inclusive. One of Cohen’s own former food science professors from Penn State doesn’t think the CEO has quite grasped a full understanding of the field: “There’s a wisdom-of-the-crowds angle here that most people haven’t leveraged in sensory [science],” says John Hayes, PhD. “But the idea that you have to throw away calibration, or good sensory practices, to get the scale [of data] he’s talking about — I just don’t think that’s true.”
And with all the current hand-wringing over digital privacy in the wake of Cambridge Analytica, it will also remain to be seen just how many people are eager to submit this kind of incredibly detailed data in exchange for more appealing flavors of beer or potato chips.
Big food manufacturers already rely on something called “sensory science” to develop addictively crunchy potato chips or the perfect cookie, but the development of such products relies largely on collecting data from consumer testers — and people are notoriously unreliable and unpredictable. Launching a new food product is risky, costing the food industry an estimated $20 billion annually, but relying on artificial intelligence and huge quantities of data could be a more reliable way for food companies to figure out what people want to eat.
• Can A.I. Usher in a New Era of Hyper-Personalized Food? [The New Food Economy]
• Can Artificial Intelligence Brew Better Beer? [E]