AI Podcast: How AI Can Create Better Beer, Yummier ChocolateFebruary 4, 2017
Whether brewing crisp lagers or creating a new candy bar, flavor is a fickle thing. And that’s a big business problem for the food and beverage industry.
On this week’s edition of the NVIDIA AI Podcast we talk to Jason Cohen, founder of Analytical Flavor Systems, who is using AI to help businesses that create beer, chocolate, wine, coffee and spirits better understand flavor.
“Something crazy like 95 percent of all new products fail, either in the conception phase or in the R&D phase or in the go-to-market phase,” Cohen says in a conversation with our host, Michael Copeland. “That’s because these companies can’t predict consumer preference, because they don’t have a measurement of what consumers are going to taste in that product.”
Analytical Flavor Systems uses GPU-powered artificial intelligence to turn data about what people taste — whether it’s beer, wine, chocolate, wine, coffee and spirits — into data companies can use to create more consistent — and tastier — products.
That’s just as important in the food industry as it is for more traditionally data driven industries. “Food and beverage companies don’t have a map of consumer preference, and consumer preference is constantly changing for what people want to taste,” Cohen says.
Case Study: How IPAs Blindsided Big Brewers
That kind of map can help businesses predict — and get ahead of — major disruptions, such as the shift to a once obscure breed of bitter brews known as India pale ales, or IPAs.
“If you think back 10 years ago, no one drank IPAs. IPAs were not a market-winning style. And then suddenly there was a global preference shift, and IPAs went from negligible — not even 1 percent of global beer sales — to now being a leading style category,” Cohen says.
The result: some of the world’s largest brewers missed out on being market leaders in IPA, giving that up to the craft brewing industry.
“Being able to predict where preferences are going can lead to new, successful products and better tasting products for everyone,” Cohen says.
How Humans Bias AI (and How AI Might Help Us Be Less Biased)
And if you missed Episode 7 of the AI Podcast it’s worth a listen: Narrative Science Chief Scientist Kris Hammond explains how our own biases can creep into the AI systems we build — and how AI may help us one day get past our own cognitive blind spots.
Featured image: Quinn Dombrowski