How AI will impact product development
Advances in artificial intelligence in the food and beverage industry will push boundaries, elevate innovation, and improve success rates of new product introductions.
Developing recipes that create craveable foods and beverages has always been a delicate balance of art and science. Chefs, food scientists and other professionals rely on intuition, craft and an understanding of the compounds that interact to create flavor. The future, however, may be far more science based as artificial intelligence (AI) is being developed and increasingly employed to not just predict flavor trends, but also to develop flavors that may become trendy.
Chefs and other culinary professionals are frequently trying to push boundaries and create the next trend-worthy creation that stands apart from what’s been done before. But those efforts are often bound by experience, preferences and inherent biases based on culture, training, and expectations. AI suffers from none of those expectations or limitations, which makes its entry into the flavor scene as exciting as it may be as concerning to some.
One of the earliest entrants into this field has been IBM’s Watson. After winning on Jeopardy!, the team has partnered with the Institute of Culinary Education in New York City to teach the system to better understand flavor and design. They’re working on developing unique recipes that break flavor boundaries and create combinations that are both appealing and completely new. Watson has already partnered with Nunu Chocolates to create novel combinations and custom recipes for the company’s chocolate truffles. Callebaut used AI to better understand unmet consumer preferences and tastes which led to the development of its ruby chocolate, launched in 2017.
Gastrograph AI may not create new flavors but it does help understand the appeal of the flavors created and optimize novel flavors for target audiences, all using AI.
Developing flavor using AI isn’t just happening in the kitchen to develop final products. Fifth Season, a vertical farm in Pittsburgh, is combining big data and AI to create the optimal growing “recipe” that will determine the best flavor for each plant.
It’s important to note that the art hasn’t been abandoned in favor of science. For example, all the recipes created by Watson in collaboration with the Institute of Culinary Education are tested by chefs. As with any culinary endeavor, there are hits and misses, and that information is fed back into the system to improve the outcomes. In the end, AI may be able to predict craveability but humans will have the ultimate word on how successful those efforts are.
Technology is inevitable in modern markets, and the food and beverage industry will not be exempt from this evolution. AI will undoubtedly become a larger part of the development process to more effectively create and refine flavors, capitalize on white space, and increase the potential success of new product releases.
Consider the complexity of launching a beer product, with its heavily nuanced flavor profile amid heavy competition. Carlsberg is now investing millions of dollars in AI-based research that can detect the nuances of taste in its new beer and optimize those flavor profiles for its target marketplace.
It’s important to remember that flavor appeal can be impacted by non-flavor attributes such as texture. Ingredion Inc has used AI in its T-Rex robot for over 10 years now, and this robot has led to breakthroughs in texture that improve the overall eating experience of a variety of products. Though the excitement may be around the completely new flavors that may be created, the reality is AI can save or earn a company millions of dollars by creating products that will get to market faster and be more successful after launch. That is the impetus that is sending companies like Danone, Kellogg’s, Cargill and Dole in search of AI supplies such as Ai Palette. AI will be a key element to the future of the food and beverage industry and its continued success in an ever-evolving modern marketplace