(Translated by https://www.hiragana.jp/)
Michael Wolf
Close Ad

The Spoon

Daily news and analysis about the food tech revolution

Michael Wolf

Anova Serves Up a Generous Helping of AI With Launch of Anova Precision Oven 2.0

Last week, Anova announced the second generation of its Precision Oven, just over four years after it began to ship the first generation Precision Oven. The Anova Precision Oven 2.0 is packed with a number of new features, including an in-oven camera and Anova Intelligence, a suite of new AI features designed to power new ways for users to assist in the cooking process.

In fact, the company listed a bunch of features currently offered by their AI-powered cooking and a number of features that they are working on. The features the Anova Precision Oven 2.0 currently has include:

  • Ingredient Recognition: The AI system automatically identifies what’s inside the oven with the internal camera.
  • Suggested Cooking Methods: The oven’s AI will suggest cooking methods tailored to the ingredients, ranging from basic roasting to more complex sous vide style.
  • Packaged Food Conversion: The oven will scan the packaging, and the AI will choose the right settings.
  • Recipe Conversion: The company says the Anova Precision Oven 2.0’s AI can convert nearly any recipe to make it work with its settings, with the caveat that this feature will improve over time as it gains more data.

According to Anova, upcoming features for Anova Intelligence include:

  • Assistant Mode: Anova’s AI-powered co-pilot will simplify complex cooking techniques and offer personalized cooking guidance.
  • Complex Meal Creation: When preparing multi-component dishes, the oven will suggest optimal settings for each ingredient, “streamlining” the cooking process for recipes that typically require juggling multiple cooking techniques.
  • Cook Recall: The oven will recognize repeat recipes for dishes you prepare frequently and return to your last-used settings.
  • Doneness Detection: Powered by the internal camera, the oven monitors the cook’s desired crispness level and alerts them when it reaches the preferred doneness level.
  • Auto Shutdown: The oven will detect when a cook cycle has finished and whether food has been removed, then notify you before automatically shutting off.
  • “Clean Me” Reminders: Equipped with an internal camera that monitors for dirt buildup, the oven will remind you when it’s time for a cleaning.

With the addition of the camera and new AI features, it looks like Anova hopes to fill the void left after Weber sunsetted the June oven about a year ago. While some features (like auto-shutdown) don’t seem all that interesting, I am intrigued by features like the coming co-pilot mode.

In addition to the new AI features, the second-gen oven includes even tighter temperature management (powered by three internal temperature sensors and a more powerful on-board processor) and better steam management. The new oven also includes a new app and an additional recipe subscription service for $1.99 a month or $9.99 annually.

All of these new features come with a hefty price tag increase at $1199, double that of the launch price of the first-gen oven. While some may pass on the 2.0 due to the price increase, given the void left by June and the cult following Anova has, I expect the new Anova oven to sell fairly well when it ships.

Lee Kindell on His Journey from Hostel Operator to Building a Fast-Growing Robot-Powered Pizza Chain

Lee Kindell, founder of MOTO Pizza, Seattle’s fastest-growing restaurant chain, never intended to be a restaurateur. In fact, starting a small local pizza joint was more of a retirement plan for when he eventually retired from running a hostel, where he’d cultivated a community atmosphere by taking guests on foraging trips, introducing them to local food, and hosting communal dinners.

According to Kindell, it was during one of these gatherings that someone suggested making pizza, and before long, he became fascinated with the craft of dough preparation. Over time, he learned the nuances of sourdough and grew to love the tactile joy of working with dough, and he would talk to his wife, Nancy, about an eventual far-away plan to retire and run a small pizza restaurant.

But like millions of Americans, the COVID-19 pandemic pushed him into a pivot faster than he anticipated when his hostels closed, and his far-away plan soon became the next thing up. He poured his savings into a 500-square-foot space in Seattle’s West Seattle neighborhood—which they affectionately call “The Up House.” With less than $60,000, Kindell bought used equipment and salvaged furniture from the side of the road, and soon, he soft-launched his little pizza shop. But, in a shock to both Kindell and his wife, a soft launch quickly turned into a viral sensation as bored home-bound customers in the midst of a pandemic lined up around the block after learning about MOTO on social media and from an article in Seattle Eater.

“We went viral on day one,” Kindell said.

From there, MOTO’s reputation grew as Kindell dedicated himself to bettering his craft. He became a hands-on pizzaiolo, mixing dough by hand until an arm injury pushed him to adopt a mixer—a pivotal moment that reshaped his views on scaling. Kindell realized that the mixer didn’t compromise the dough’s quality, and soon, he began exploring other tools, such as conveyor ovens and robotics, to do what he called “scaling craft.”

Moto’s expansion soon took an unexpected turn, landing a coveted spot at the Seattle Mariners Stadium after one day talking to a customer who worked for the team. Kindell pitched his idea to make stadium-friendly, hand-held pizzas, and it wasn’t long MOTO was serving up to 1,400 pizzas per game. Success at T-Mobile Park led to additional opportunities, including the Paribas Tennis Open in Palm Springs, where MOTO served 3,000 pizzas a day during the tournament. “It’s a logistical nightmare,” Kindell said, but he is confident in scaling MOTO nationwide without losing the craft and authenticity that made it special.

Nowadays, MOTO continues to grow (they just opened their first mall location at Bellevue Square) and Kindell has plans to expand nationally and even overseas. There’s no shortage of investor interest, and Kindell and his team plan on taking on more investment next year as they expand their footprint and explore new technology options like drone delivery.

You can watch our full conversation below or listen to it on Apple Podcasts, Spotify, or wherever you get your podcasts.

MOTO Pizza's Lee Kindell on taking chances and making pizzas [VIDEO]

The Idea of Food ‘Teleportation’ Isn’t New—But AI Is Finally Making Distributed Digital Food Replication a Reality

Over the past decade, there’s been no shortage of attempts to better understand, map, and recreate food and its properties with the click of a mouse.

From new data ontologies to a proliferation of digital noses, we’ve seen incremental steps toward an envisioned world where the fundamental building blocks of food can be better understood. However, in the past year, there has been a rapid acceleration in our collective ability to digitize various properties of food, largely driven by advances in AI.

The latest example of this comes from Osmo, which recently announced its development of the ability to digitally “teleport” a scent by using AI to digitize and re-materialize it.

Scent Teleportation

As the company’s CEO, Alex Wiltschko, explained:

“We select a scent to teleport and introduce it to a machine called the GCMS (Gas Chromatography-Mass Spectrometry). If it’s a liquid, we inject it directly; if it’s a physical sample, like a plum, we use headspace analysis, trapping the scent in the air around the object and absorbing it through a tube. The GCMS identifies the raw data, which can be interpreted as molecules, and uploads it to a cloud. There, it becomes a coordinate on our Principal Odor Map — a novel, advanced AI-driven tool that can predict what a particular combination of molecules smells like. This formula is sent to one of our Formulation Robots, which treats it as a recipe and mixes different scents to replicate our sample.”

In other words, Osmo breaks down the building blocks (molecules), creates a map, and then sends this digital map to an essence “printer” that re-creates it.

This announcement comes just weeks after leaders of NotCo’s scent and flavor AI team shared research on their new generative AI that creates scent and flavor formulations. Here’s how Aadit Patel, NotCo’s head of product, described the model:

“The system takes your prompt—such as ‘an ocean scent on a breezy summer day on a tropical island’—and creates a novel chemical formulation of that scent in one shot.” The model then generates a corresponding fragrance formula. According to Patel, the model is built on a “natural language to chemical composition” framework, tokenizing molecules to create a system capable of understanding and generating novel combinations.

With years of work focused on digitally understanding, quantifying, mapping, and reproducing scents, flavors, and food building blocks, what is allowing these latest efforts to make such significant leaps?

In a word (or two): AI. In the past, creating a facsimile of a flavor or scent took thousands of hours, relying on trained experts to work in a lab or kitchen, drawing on years of expertise. Now, AI is expediting that process with orders of magnitude more efficiency, often leaving the expert to provide a final sign-off to ensure the AI-created formula meets standards for proximity to the desired result, as well as checks for safety, cost feasibility, and more.

For the record, Osmo isn’t the first company to discuss “teleporting” a formula for digital recreation. In 2018, Japanese startup Open Meals made headlines with its “sushi teleportation” demo, essentially sending 3D printing instructions to create a sushi-like meal. We also saw Cana’s ambitious attempt to make a Star Trek replicator (though, as it turns out, investors weren’t quite ready to enter the food teleportation age).

All of this follows years of efforts to quantify and understand food digitally, including the creation of ontologies for the Internet of Food and early attempts to use AI to analyze food. But over the past couple of years, there’s no doubt that parallel advances in AI (especially in large language models) and breakthroughs in food, olfactory, and chemical science are ushering in a world where true food “teleportation”—or, more accurately, the ability to understand and synthetically recreate food, flavors, and scents—has arrived.

I’m excited to see where this all goes. To manifest the vision laid out in science fiction over the years and imaginative product concepts like that of Open Meal required a true digital understanding of the molecular building blocks of food. With AI, we are closer than ever to that understanding, and the products we’ll see built in the coming decade will not only create some mind-blowing consumer experiences but also possibly fundamentally change how food and beverage products are made and distributed.

Brava’s Latest Update Enables Blind and Low-Vision Users to Cook Independently With The Light-Powered Oven

This week, Brava, a company that makes countertop consumer appliances that cook with light, has announced the launch of the first major update to the software since 2021.

The biggest change announced is the ability to fully access and control the Brava through the app. Unlike previous versions, which required some interaction with the Brava’s touchscreen, users can now manage ingredient selection, doneness, and recipe instructions entirely through the Brava app, reducing reliance on the oven’s interface. This update is particularly impactful for those with physical limitations, allowing them to enjoy the Brava’s convenience and precision with ease.

While full app control of an appliance is something some folks make like (and others may hate), what’s neat about this update is that Brava chose to also enable new features that make the Brava more accessible to individuals who are blind or have low-vision.

As can be seen in the video below, the key advancement hear is the Brava app works with screen readers, which are digital assistants that work with apps that allow those that are blind or have low vision to navigate.

“Screen readers so it changes the behavior of your touchscreen device, “said Fernando Macias, Access Technology Specialist at San Francisco Lighthouse for the Blind and Visually Impaired. “So that everything you touch on the screen, (the screen reader) will read out loud, and that is what enables us to navigate, especially if you’re a blind or have have low vision.” Brava worked with San Francisco Lighthouse to enable the screen reader features.

In addition to greater accessibility features for low-vision users and full control of the appliance via the app, the updated software provides users with more granular info on light intensity and duration parameters that drive each recipe, empowering them to modify or create recipes with greater insight.

Long-time Spoon readers may recall that The Spoon was the first press org to get a look at the Brava back in 2018. Just over a year after the Brava was introduced to the work, Middleby acquired the company. Interestingly, Brava’s founding team (Thomas Cheng, Daniel Yue, and John Pleasants) are all still with the company five years after the acquisition, a sign that Middleby continues to invest and show interest in the platform.

A First Look at Roku Shoku, Sony’s Culinary Recording System to Capture and Replicate Chefs’ Recipes

This past week in Japan, Sony unveiled a project they’ve been developing in secret called Roku Shoku, a culinary recording system designed to capture exactly how a chef prepares a meal. The system also serves as a guidance tool, helping casual or inexperienced cooks create dishes with the precision of an expert with years of training.

Sony has been working on this project, which stands for “Record” (Roku) and “Cooking” (Shoku), for the past five years. Last week, the entertainment and consumer electronics giant held the first-ever press demonstration of the recording studio for The Spoon team.

“We have a recording studio here in Tokyo,” said Tomoko Nomoto, Project Leader for Roku Shoku. “We invite Michelin-starred chefs, or even grandmothers, to the studio and ask them to cook with our system. We then record their culinary data, including temperature, steam levels, and the entire cooking process.”

The project is led by a Sony R&D team out of Tokyo and is separate from research in the area of gastronomy that has taken place at Sony’s AI Research division Sony AI (the formal Gastronomy program announced in 2020 has been sunsetted, but Sony continues to work on gastronomy-related projects). Since launching the Tokyo recording studio in 2021, the team has captured thousands of recipes across a range of cuisines, including Japanese, Chinese, French, Italian, and Thai.

The Roku Shoku system features induction cooktops with temperature sensors, scales to monitor and weigh ingredients, cameras to capture a chef’s movements, and an off-the-shelf game controller made by Steam to control the setup.

Nomoto shared that users can replicate meals precisely as chefs cook them, a claim I tested myself. You can watch me trying it in the video below.

First Ever Look at Sony's Roku Shoku Culinary Recording System

According to Nomoto, the goal is to use Roku Shoku both to document recipes for restaurants and food service locations and to preserve culinary creations for future use.

“The first step will be to work with restaurants that want to share a consistent experience worldwide or recreate dishes that are no longer available, like when a chef passes away or retires,” said Nomoto.

Spoon readers might recall Cloudchef, another system that records chef creations. Nomoto explained a key difference: Sony plans for Roku Shoku to enable only human chefs to recreate these meals, while Cloudchef eventually aims to use robots for meal replication. Currently, both systems are focused solely on human use (see Spoon’s Tiffany McClurg using the Cloudchef system here).

Sony plans to roll out Roku Shoku in Japan in early 2025, with expansion to other regions to follow. The company has launched a website where you can find out more and request a demo.

Three Days of Food Innovation & Discovery in Tokyo: Scenes From SKS Japan 2024

In January of 2017, little did I know that a chance meeting in the Venetian in Las Vegas would eventually lead to the creation of Japan’s most influential and well-attended food tech event.

We’d just come off our second North America Smart Kitchen Summit a couple of months prior, and two first-time attendees who had made the trip to Seattle, Hirotaka Tanaka and Akiko Okada, asked if we could meet during CES.

I said yes, and the rest is history. Half a year later, we got together in Tokyo for the first-ever SKS Japan, and I was wonderfully surprised at the excitement and innovation around the food system taking place in this beautiful country. And now, seven years later, we just had the seventh SKS Japan, and it’s incredible how this show has matured into one of the most important food tech events on the calendar.

This year’s event spanned three days and brought in over one hundred speakers from across the globe and the food system. Below are some snapshots from the show.

Above: Myself and Hiro Tanaka giving the opening keynote to kick of SKS Japan 2024 (and announce SKS 2025 dates).

Above: Robin Liss (Suvie), Kevin Yu (SideChef) and Assaf Pashut (Chefee) join me for a conversation about how the consumer kitchen has evolved over the past decade and where it will go over the next ten years.

Above: Jasmin Hume (Shiru) joins me on stage and Adam Yee (Sobo Foods) and Tarini Naravane (Grainge AI) dial in to talk about the impact of AI on developing new food inputs and products.

Above: Lee Kindell, the founder of MOTO Pizza, comes to Japan from Seattle to talk about his fast-growing restaurant and how he leverages technology such as robotics to help “scale craft”.

Above: Author and Wired’s kitchen tech reviewer Joe Ray sits for a fireside chat on his thoughts about food tech journalism and first impressions of Japan’s food and restaurant scene.

Above: A talk about new approaches using upcycling and food life extension technologies with Moody Soliman (Ryp Labs), Daichi Takada (AlgaleX Co.), and Lina Sakai (Fermenstation).

Above: Akiko Okada (UnlocX), Hirotaka Tanaka (UnlocX), Michiaki Matsushima (WIRED JAPAN), Hiroki Nakajima (University of Tokyo) during a session entitled Tokyo Regenerative Food Lab in Action

Above: The SKS Japan exhibition space. Bottom right is demo table for AlgaleX’s upcycled algae protein product line, Umamo.

Above: Exhibitors head out to the streets of Tokyo for a street fair on the final day of SKS Japan to share their innovations with a wider audience.

Meet The Reimagining Restaurants Podcast, and Check Out The First Episode With Wow Bao’s Geoff Alexander

Here at The Spoon, we’ve covered hundreds of restaurant operators over the past half-decade, in part because we love restaurants and restaurant tech, but mostly because restaurant operators are some of the most creative and hard-working entrepreneurs in the food business.

The truth is they have to be. There’s no business changing faster than the world of food service, with the advent of ghost kitchens, digital ordering, automation, and AI, all against a backdrop of high food costs, changing consumer tastes, employee turnover, and more.

All of which is why we’ve decided to start a new podcast focused on entrepreneurs who are finding new ways to run a restaurant in today’s modern world. We wanted to hear their journey into restaurants, hear how they are rethinking how to do business in today’s world, and where they see the restaurant business going in the future.

For our first episode, we knew no one better to discuss approaching the restaurant business in new and innovative ways than Geoff Alexander. The CEO of Wow Bao got his start in hospitality with a college job as a bartender, where he discovered his talent for connecting with people, and he hasn’t looked back.

He joined Lettuce Entertain You shortly after graduation and rose through the ranks over the next three decades, gaining operational expertise and eventually overseeing his own division. In 2009, he took the reins at Wow Bao, where he embraced technology as a way to turn the business around and innovate on new business models. He introduced self-ordering kiosks, mobile ordering, and partnered with third-party delivery platforms early on, which laid the foundation for Wow Bao’s unique model that spans over 500 locations with a mix of virtual kitchens and centralized food production.

Alexander shares insights on Wow Bao’s approach and how a lack of capital fostered a culture of creativity and efficiency. This approach led to successful innovations like centralized production and distribution, which he says has kept cost low while ensuring quality. Alexander also discusses scaling with hot-food vending machines and dipping the company’s toe into the metaverse.

It was a fun episode, and I think you’ll enjoy it. Listen and subscribe on Apple Podcasts or wherever you get your podcasts!

GFI: Fermentation Fuels Q3 2024 Alt-Protein Investments, While Cultivated Meat Lags

This week, the Good Food Institute released its quarterly alt-protein investment update, so I thought it’d be a good time to check in and see which way things are trending.

At this point it’s well-known that the alt-protein investment climate has had a couple tough years, and the latest numbers show things remain tough.

According to data shared by GFI, alternative protein companies raised $233 million in Q3 2024—a 37% decrease from Q2, but a 25% increase year-over-year. The drop from Q2 is notable, exceeding the 20% decline across the broader venture capital market. However, the 25% year-over-year growth in Q3 is a potential bright spot and could indicate that the second half of 2023 marked the low point for the overall market.

Source: GFI

GFI also broke things down by sub-sector, and it’s clear that fermentation technologies are where investors are placing their bets:

  • Plant-based proteins raised $56 million in Q3, bringing the year-to-date total to $194 million.
  • Fermentation technologies, which saw the largest share of investment, raised $174 million in Q3, with $572 million invested year-to-date.
  • Cultivated meat and seafood companies raised $3 million in Q3, reaching $133 million year-to-date.

It’s important to note that the overall alt-protein investment numbers can be heavily influenced by a few large deals. For example, Q2 2024 saw $118 million invested in the cultivated meat sector, largely driven by a $55 million Series B investment in Prolific Machines, a cultivated meat infrastructure company, and a $42 million investment in Dutch cultivated meat pioneer Mosa Meat. These larger deals led to an average deal size of $10 million for cultivated meat in Q2, compared to a paltry $396,000 in Q3 2024.

Similarly, Q3’s fermentation investment numbers were significantly impacted by two major deals: a $61 million investment in fermentation startup Formo for its Koji cheese products and a $45 million Series B for precision fermentation startup Helaina, focused on its human lactoferrin product.

GFI notes that lower interest rates moving forward could provide a boost to the alt-protein space, but cautions that the cost of capital remains relatively high. They continue to advocate for alt-protein startups to explore non-traditional funding sources, such as government-backed loans and programs.

Looking ahead, I predict that fermentation-based startups will remain the most attractive area for investors in the coming year. Investment in cultivated meat startups will likely focus on infrastructure players with game-changing technology, like Prolific Machines. Meanwhile, many cultivated meat startups that raised significant rounds in the past few years to scale manufacturing have put those plans on hold as they work to extend their funding runways during this ongoing VC winter.

According to GFI, the alt-protein space has seen a cumulative $16.3 billion invested since 2015, a decent number overall but still relatively small compared to other sustainability focused-sectors. To give you an idea of just how small the space is compared to other sectors, the solar industry raised $6.9 billion in venture capital in 2023 alone, and that number jumps to $34.3 billion when factoring in corporate funding.

One thing that the space needs to attract bigger dollars is a more attractive exit outlook. Overall, the exits in alt-protein has been disappointing, as have the results of those companies that have gone public. Until we see a big investor success story in this space, the dollars may remain relatively small compared to other markets.

Live Event: Using Generative AI to Build Next-Generation Flavors & Fragrances

Join us today for a live event at 8:30 AM PT, featuring the project leads for NotCo’s new AI to develop flavors and fragrances. Register and watch below or head to Streamyard.

Can AI be used to create new flavors and fragrances?

As I wrote last week, food-tech company NotCo has been asking itself this question for the past couple of years. Their answer is a newly unveiled generative AI model, the Generative Aroma Transformer (GAT), that they say is capable of creating new flavor and fragrance formulations.

The company’s Senior vice president of Product, Aadit Patel, described how it works this way: “The system intakes your prompt, such as ‘an ocean scent on a breezy summer day on a tropical island’, to create a novel chemical formulation of that scent in one shot.” From there, the model generates a corresponding fragrance formula. According to Patel, the model is built on a “natural language to chemical composition” framework, tokenizing molecules to create a system capable of understanding and generating novel combinations.

NotCo says early tests have been extremely positive, and the company says their research indicates that GAT’s abilities rival those of human perfumers. At the Food AI Summit last month, the two product leads, Francisco Francisco Clavero and Cindy Sigler, gave an in-depth presentation on the science behind their new model and talked about early results.

Their presentation was fascinating, so I asked them to present to our Food AI Co-Lab community.

Watch the recorded session below:

Using Generative AI to Build Next-Generation Flavors & Fragances

Robot Delivery Startup Starship Teams Up With European Food Delivery Company Bolt

Sidewalk robotic delivery Starship Technologies announced this week they are teaming up with Bolt, a European multiplatform delivery company, to launch a new food delivery service using Starship’s autonomous robots in Tallinn, Estonia.

The launch, centered in Tallinn, has the potential to reach up to 180,000 residents according to Bolt. Starship’s robots will operate from three Bolt Market locations—Tulika, Pallasti, and Mustika—via the Bolt Food app. During the launch period, customers will get to use the service for free (the companies did not disclose how much the service will cost post-launch).

Starship robots can carry up to three bags of groceries within a 3-kilometer radius. Bolt customers can opt for “robot delivery” through the app, meet the robot outside their location, and unlock it using the app to retrieve their items.

“This collaboration is not just about convenience and choice,” said Ahti Heinla, who cofounded Starship Technologies with fellow Skype cofounder Janus Friis. “Integrating our robots into Bolt’s service offers a scalable, sustainable delivery solution that reduces traffic and emissions. This is an exciting step forward toward greener cities across Europe.”

Starship, which was the first company to launch the first sidewalk delivery robot a decade ago, has completed over 7 million deliveries globally and traveled more than 14 million kilometers in 100 locations worldwide, according to the announcement. Operating at L4 autonomy since 2018, the company says its robots perform 150,000 crossings daily.

The two companies plan to expand beyond Tallin, but have yet to give a timetable for expansion. The news comes a week after Starship became the first sidewalk delivery platform approved for delivery in Minneapolis, delivering from Panda Express, Starbucks and Erbert & Gerbert’s.

Cashierless Checkout Pioneer Grabango Shuts Down After Failing to Secure Additional Funding

Grabango, a grocery-tech startup that raised over $93 million for its cashierless checkout technology, is shutting down, The Spoon has learned. The closure follows the company’s inability to secure the necessary funding to continue operations.

In a statement to The Spoon, Grabango said:

“Grabango announced today it has permanently discontinued operations. Although the company established itself as a leader in checkout-free technology, it was not able to secure the funding it needed to continue providing service to its clients. The company would like to thank its employees, investors, and clients for their hard work and dedication. The decision was an extremely difficult one to make.”

Founded in 2016, Grabango emerged during a surge of investment in grocery checkout technology startups, spurred by Amazon’s launch of Amazon Go. However, the field quickly became crowded with competitors like Shopic, Trigo, Mashgin, and Caper (acquired by Instacart), all of which offered variations of computer vision and AI-powered shopping platforms.

Despite the competition, Grabango secured notable clients, including European grocery giant ALDI, which just six months ago introduced its ALDIgo checkout-free solution, powered by Grabango’s technology. Yet, as seen with Amazon’s recent rollback of its Go platform in Fresh stores, cashierless checkout needs to be carefully deployed because customers can sometimes find their friendly cashiers being replaced by a technology platform offputting.

Grabango’s shutdown is a reminder of the tough climate for startups today. The days of easy venture capital are over, and in highly competitive sectors like grocery tech, startups that can’t extend their financial runway or achieve profitability are vulnerable. It’s likely that Grabango’s assets and intellectual property will soon be scooped up by a competitor.

NotCo Has Created A Generative AI for Flavor and Fragrance That Can Create Unique Formulations With Text Prompts

Food-tech company NotCo has developed a novel generative AI model, the Generative Aroma Transformer (GAT), capable of creating new flavor and fragrance formulations. The model, which the company presented on at the Food AI Summit last month, could be a potentially disruptive new tool that could impact a variety of consumer goods markets such as food, personal care, home care, and beauty industries.

The company’s Senior VP of Product, Aadit Patel, talked about the new model in a post on Linkedin today, highlighting how GAT can translate textual prompts into unique chemical formulations. “The system intakes your prompt, such as ‘an ocean scent on a breezy summer day on a tropical island’ to create a novel chemical formulation of that scent in one-shot.” From there, the model generates a corresponding fragrance formula. According to Patel, the model is built on a “natural language to chemical composition” framework, tokenizing molecules to create a system capable of understanding and generating novel combinations.

NotCo says early tests have been extremely positive. Their research indicates that GAT’s abilities rival those of human perfumers. In blind smell tests, fragrances created by GAT proved indistinguishable from those crafted by human experts. This finding signals a potentially significant shift in the industry, where only 600 certified perfumers exist globally.

How it Works

According to research presented at the Food AI Summit, the core of GAT’s functionality lies in its ability to understand and model complex interactions between volatile molecules. The model is trained on an extensive dataset of historical fragrance formulations and the molecular structures of volatile compounds. This training enables GAT to decipher the subtle relationships between different molecules and predict how they will interact to create specific aroma profiles.

The model employs a dual-system transformer network comprising an encoder and a decoder. The encoder processes the user’s prompt (with inputs such as top note (cherry candy), middle note (vanilla) and bottom note (cherry)), capturing the desired aroma profile. This is passed to the decoder, which generates a corresponding sequence of tokens representing the fragrance’s molecular structure.

GAT leverages the atomic structure of volatiles to generate novel formulations. Each molecule is represented as a graph, with atoms described by valence, degree, hydrogen count, hybridization, formal charge, and atomic number. These details are then translated into numerical representations and fed into a Graph Neural Network (GNN) model, which creates a unique vector representing each molecule. Similar vectors indicate similar molecules, allowing GAT to identify and utilize molecular structures with desired aromatic properties.

The potential impact of NotCo’s GAT is substantial. Developing new formulations for flavors and fragrances has traditionally been a time-consuming and resource-intensive process, often requiring weeks or months of expert work. If GAT can achieve the same outcome in mere seconds, it could significantly reduce flavor and fragrance development costs.

If you’d like to learn more about NotCo’s new generative AI tool for developing flavors and fragrances, NotCo’s head of machine learning, Francisco Clavero, and one of their key flavor and fragrance scientists, Cindy Sigler, will be our guests at the next Food AI Co-Lab on October 17th. You can register for this virtual event here.