Is your standing rib roast missing a little something, but you’re not sure what? Perhaps artificial intelligence (AI) can help. A new research collaboration between McCormick & Company, the supermarket seasoning giant, and IBM could bring science to the kitchen that is more advanced than the chemical reaction between baking soda and vinegar.
Using IBM Research AI for Product Composition, product developers at McCormick & Company will use AI to devise new flavor combinations and help predict what flavor will be trending soon. The proprietary AI will cull from “hundreds of millions of data points across the areas of sensory science, consumer preference, and flavor palettes” that was collected by McCormick over decades of past product formulas and research, as per a press release. According to IBM, the algorithms can suggest substitute raw materials (like allspice instead of cinnamon), devise novel flavor formulas, and even predict human response.
“McCormick’s use of artificial intelligence highlights our commitment to insight-driven innovation and the application of the most forward-looking technologies to continually enhance our products and bring new flavors to market,” says McCormick’s chairman, president, and CEO, Lawrence Kurzius. “This is one of several projects in our pipeline where we’ve embraced new and emerging technologies.”
After a four-year joint effort, McCormick expects to launch the first AI-enabled product platform, ONE, by mid-2019, with a set of initial one-dish Recipe Mix flavors including Tuscan chicken, bourbon pork tenderloin, and New Orleans sausage. The ONE platform was specifically developed to create new seasoning blends that can “season both the protein and vegetable,” which they expect to be on U.S. retail shelves by late spring.
Hopefully things will get as weird as that time when a neural network was asked to name ice cream flavors and decided “Peanut butter slime” and “strawberry turfle” were good options.