What’s Your Flavor?

Personalizing food with the power of AI

What does the future taste like? Coca-Cola’s betting it tastes like Y3000, created with the help of AI to “bring the flavor of tomorrow to Coke fans.” With the fast-processing abilities of AI, Coca-Cola and other companies are also betting on being able to personalize that taste to individual consumers. Using a combination of data from consumer preferences, real-time trends, chemistry, and sensory profiles, AI can discover new ingredients and flavors, vastly reducing the amount of time it takes to formulate new products. And it can optimize those products to taste better while being healthier.

Viral Flavors

Coca-Cola’s not alone in rolling out new products and flavors with the help of AI. This year, Baskin Robbins Korea has debuted two AI-generated flavors: Orange Earl Grey developed with ChatGPT and Tropical Summer Play designed by Google Gemini.

“Through this project, we have confirmed that there is a substantial possibility to employ next-gen AI into our new product development,” a company official said. “This unique combination of technology and innovative ideas will offer customers not just delicious products but also special and differentiated experiences.”

Baskin Robbins chose the color scheme for the new flavor and asked Gemini to choose ingredients based on those and search terms associated with summertime. The launch event for Tropical Summer Play also featured Gemini recommending flavors to customers based on their moods and personality types. This customization speaks to younger generations that want products tailored specifically to them.

“Major brands are listening to this powerful demographic and are moving fast to launch products that have been shared and gone ‘viral’ on social media,” Fran Padgham, marketing manager at flavor formulator Lionel Hitchen, told Food Ingredients First.

The marketing possibilities for unique flavors and creations are wide open.

“Young consumers are craving bold and unusual flavor combinations, driven by social media’s influence and a desire for novelty,” Ian McGarvey, senior VP for global technology and innovation at Kerry, told Food Ingredients First. “This opens opportunities for sweet-savory pairings like bacon milkshakes, coffee infused with black garlic, and chocolate bars with wasabi.”

Fast Food

While around 80% of new food and beverage products fail to gain a strong following among consumers, food scientists are hoping that AI and machine learning can improve uptake, or at least help them fail faster.

“For me, the whole digitization agenda is so relevant and so exciting because, if done well, it should truly move things much faster,” Miriam Überall, a former R&D executive at Kraft Heinz and Unilever, told AgFunder News. “Avoid a lot of tests and trial and error approaches which a classical R&D organization still has and go into predictability at a much faster pace.”

By analyzing the texture and aroma of different foods to optimize palatability, AI can help ensure new products are likely to be embraced by the target audience. Using real-time trend data, it can recommend when to launch products for maximum acceptance and help food companies respond quickly to shifts in the market.

“We know from AI predictive modeling by our applications team that we can create category specific formulations,” Greg Aloi, VP of Ingredion’s Customer Co-Creation and Applications organization, told BOSS. “If we’re starting a journey in a 10-step stairwell, predictive formulations modeling will enable us to start on the fifth step instead of the first step. We are also looking to deploy AI in other avenues of the company and have several projects that will shape the way in which we interface with customers and consumers in the future.”

As with the low-salt bullion Unilever developed using AI, the tech can improve nutritional content without sacrificing taste, in a matter of days.

Specialized Diets

AI’s flavor enhancement and ingredient discovery capabilities are being put to the dietary preferences test as well. Kraft Heinz teamed with AI-powered NotCo to produce NotMayo, a mayonnaise alternative that uses chickpea flour and other plant-based ingredients. Multinational cheese marketer Bel Group partnered with Climax Foods to use its SuperFood Intelligence to make plant-based versions of its cheeses.

Firmenich developed what it bills as the world’s first AI-created flavor to provide a natural beef taste for plant-based meat. “The flavor was created using a rule-based formula generation model leveraging raw material usage statistics, spanning the entirety of Firmenich’s broad formulae databases,” the company said. “The flavor capitalizes on Firmenich’s unique palette of ingredients with a delicious and complex meaty, fatty, long boiled, and slightly grilled flavor profile.”

In the case of lab-grown meat, companies like Animal Alternative Technologies aim to help the industry scale by using AI to determine and maintain optimal growing conditions. It’s all about enabling consumers to meet their dietary requirements without sacrificing flavor. That’s the taste of the future.