Care for a sample of Eau de AI? One day you might if the partnership between IBM and global fragrance company Symrise transforms perfumery from a bespoke art to a matter of algorithms. Vox reports that Symrise, whose clients include Estée Lauder, Avon, Coty, and Donna Karan, sought the expertise of IBM to explore machine learning opportunities for the business.
IBM’s answer is Philyra, an algorithm using existing fragrance formula data as well as demographic data such as geography and customer age. Philyra’s powerful market segmentation capabilities have already been applied to two AI-developed perfumes to be sold at O Boticário, the second largest beauty retailer in Brazil. The scents will be on shelves in 4,000 locations next year.
Symrise’s impressive collection of 1.7 million fragrance formulas - ranging from consumer-facing scents to household scents like snacks and detergent - along with accompanying customer and sales data were added to Philyra’s database.
Its ability to perfect fragrance combinations through machine learning makes Philyra “like a human apprentice,” per IBM researcher Richard Goodwin.
"It’s a very human reaction to be concerned about technology replacing a human, and we’re not talking about replacing perfumers tomorrow."
For the O Boticário scents, Philyra was asked to find scents that would resonate with Brazilian millennials and came up with two, one more exotic “with a milky, buttery, rich base note,” and the other “a fruity, floral scent,” according to Symrise senior perfumer David Apel. Both scents enjoyed rave reviews from focus groups.
Symrise executive Achim Daub first reached out to IBM, impressed with Chef Watson’s culinary skills in a partnership with Bon Appetit and curious to see whether its ability to concoct creations could transfer from food to fragrance. His hope is that AI can draw connections that humans would otherwise overlook and target “missed opportunity” in fragrance sales.
Human perfumers are not getting the pink slip anytime soon. “It’s a very human reaction to be concerned about technology replacing a human, and we’re not talking about replacing perfumers tomorrow,” clarifies Daub. “This isn’t something to necessarily market to customers, but the internal capability is that can increase efficiency and speed.”