Coca-Cola unveiled its new limited-edition Year 3000 flavor recently and it has been all the hype in the generative AI world for two reasons. Firstly, it was “co-created” using generative AI. Secondly, it was a symbolic approval from the godfather of all F&B companies in the world for generative AI.
Both the packaging and flavor of Coke Y3000, the name of the new flavor, were created with generative AI. The design of the Coke can was generated using the image-generative AI Stable Diffusion despite the ongoing lawsuit with media company, Getty Images, over the AI being trained on copyrighted content. However, Coca-Cola is unfazed by this controversy, seeing that Coke Y3000 is already available to consumers.
The more interesting concept is how the flavor of Coke Y3000 was co-created with AI. Coca-Cola wanted to “explore the concept of what a Coke from the future might taste like”. While full AI mixologists are still at least a decade away, the company relied on human insights to determine what flavors were associated with the future and used AI to figure out flavor profiles that could pair well with these insights.
This sparked an intriguing thought: Could we be on the cusp of a future where AI crafts not just irresistible flavors, but also nutritionally optimized food and beverages? Two key factors suggest we might be.
Flavor and Smell Can Be Digitized?!
Now, if you're going to get an AI to understand food, it needs to get its virtual head around taste and smell. For a long time, smell was the Rubik's Cube of the sensory world. But scientists at the University of Reading have cracked it.
“Vision research has wavelength, hearing research has frequency—both can be measured and assessed by instruments. But what about smell? We don't currently have a way to measure or accurately predict the odor of a molecule, based on its molecular structure.”
-Professor Jane Parker, University of Reading
In a significant development, researchers at the University of Reading have engineered a tool that can predict the odor of a molecule based solely on its molecular structure. This is no small feat. Prior to this, the relationship between a molecule's structure and its odor was about as predictable as a roll of the dice. There were countless exceptions where the two simply didn't align, leaving scientists scratching their heads. However, this new AI model has managed to bring order to this chaos. It can accurately predict the odor of molecules, even those that have historically been the exceptions to the rule.
The AI model runs on an 'odor map,' a sort of digital atlas for scents. This isn't just a library of known smells; it's a framework that can also identify odors that haven't even been categorized yet. This means that the AI can potentially identify unknown odors and even assist in the creation of millions of new odorants. The AI was able to match the scent scores of a panel of humans in a survey.
Meanwhile, there's a separate but equally intriguing journey into the realm of taste. Companies like IBM are tinkering away at the future of flavor. Their AI-assisted "e-tongue" technology, whimsically named Hypertaste, has shown a knack for distinguishing between different types of mineral water with greater accuracy than human tasters. The technology works by digitizing the chemical composition of a liquid and then cross-referencing it with a database. While still in the experimental stage, the potential applications in quality control and product innovation are quite promising.
So, what does all this mean for the future? Well, once we manage to turn flavor and smell into data, we'll be stepping into a new era. Future AIs will have the capability to understand what flavors and smells are most appealing to humans. This could lead to the development of foods and beverages that are not only delicious but also nutritionally balanced. The food and beverage industry is on the cusp of a transformation, one where AI could very well be crafting the next Michelin-starred recipes. It's a future where technology and the culinary arts are set to merge, redefining the boundaries of what's possible in taste and nutrition.
Generative AI is Setting the Foundation for an F&B AI that Redefines Food and Nutrition
The underpinnings of tomorrow's culinary landscape are being laid today, and the architects are none other than generative AIs. You see, beyond the realm of GPT-4 chatbots, which are popping up in every conceivable app from meal planning to recipe suggestions, there's a more specialized breed of AI taking shape. These are generative AIs with a very particular set of skills.
Take Palette AI's FoodGPT, for instance. It's an AI trained not just to chat, but to provide nuanced insights into the F&B industry. It's like a sommelier, a chef, and a market analyst all rolled into one digital entity. FoodGPT delves into food-focused data to offer trends on what flavors are capturing the public's imagination, what types of cuisine are setting social media ablaze, and so on. The aim is to assist companies in crafting their next big hit in the F&B sector, be it a dish, a drink, or something in between.
Now, it's worth noting that FoodGPT and its ilk are still in their formative years. They're not yet capable of understanding the subtleties of flavor and smell, but they're getting there. They're laying the groundwork for a future generation of AIs that will not only understand these complexities but also use them to redefine food and nutrition as we know it. It's a future that promises to be as intriguing as it is delicious, where technology and culinary expertise converge in a most appetizing way.
Michelin Stars or Machine Learning Stars?
As we stand on the precipice of a new culinary era, it's clear that the role of generative AI in shaping the future of food and beverages is not just a fanciful notion but an impending reality. With advancements in digitizing taste and smell, we are nearing a point where AI won't just suggest what to cook for dinner but will actually design the meal itself. And not just any meal, but potentially a Michelin-starred culinary experience.
Imagine a future where your personal AI sommelier knows not just every wine in the cellar but every grape on the vine, where your digital chef understands your palate better than you do. These AIs, with their deep understanding of both the complexities of human taste and the subtleties of flavor and aroma, could craft meals that are not only extraordinarily delicious but also precisely tailored to individual nutritional needs.
So, as we look forward to this brave new world of gastronomy, one thing is certain: the marriage of technology and culinary arts is set to redefine our understanding of what is possible in the realm of food and nutrition. It's a future where the question won't be whether an AI can design a food product or a Michelin-starred recipe, but rather, how many stars the resulting meal will earn.
Comments