Close Ad

The Spoon

Daily news and analysis about the food tech revolution

News

When It Comes to Cooking Videos, Gen Zers Love TikTok, Millennials Embrace YouTube

As a former industry analyst, I’m a sucker for survey research exploring how we cook, eat, and shop for food. Luckily for me, word of a new research study landed in my inbox this morning from (of all places) Home Run Inn Pizza. Okay, so not exactly Nielsen, but the study used a good sample size (2,000 US respondents) and had a mix of gender and regional representation. In other words, it seemed to be designed well enough to elicit decent results.

The survey focused on food behavior by Gen Z and Millennials. I’d seen studies like this before – heck, we’ve even conducted them here at The Spoon – but what stood out to me about this one was just how vital the dominant video platforms are nowadays when it comes to gaining cooking inspiration. According to the survey, 71% of Gen Z (between ages 9 and 24) and 67% of Millennials watch cooking videos but differ substantially in what platforms they watch the videos on.

Source: Home Run Inn

According to the study, Gen Zers are more than twice as likely (38% compared to 16% of Millennials) to watch cooking videos on TikTok. A more significant percentage of both groups said they watch cooking videos on YouTube, but Millennials usage far outpaced Gen Z respondents (66% of Millennials compared to 47% for Gen Z). Instagram usage was surprisingly low, with only 7% of Millennials and 4% of Gen Z saying they watch cooking videos on the platform. Neither Millennials (9%) nor Gen Z (3%) watched much traditional TV when watching cooking videos. According to the survey, both generations – 56% of Gen Z and 29% of Millennials – use TikTok for recipe discovery and learning cooking techniques.

Source: Home Run Inn

Another surprising data set was the kitchen gear each used to cook food. According to the survey, both generations rely most heavily on the stovetop, with two-thirds of Gen Z and Millennials saying that was their primary appliance. Interestingly, only 10% of Gen Z and 8% of Millennials said air fryers were their go-to, and an even smaller percentage – 6% for Gen Z and 4% for Millennials – said the microwave oven was their primary cooking appliance. For some reason, the survey didn’t ask about pressure cookers, an oversight, in my opinion, despite the struggles of the pioneering Instant Pot.

Finally, a significant percentage of both generations can be scolded for being poor company when using technology while breaking bread with others. According to the survey, 81% of Gen Z admit they have stared at their phone while dining with others, compared to 60% of Millennials.

If you’d like to see the study’s full results, you can find it here.

How Mini Melts Built a $150M Beaded Ice Cream Business With a Nationwide Network of Automated Kiosks

If you’re like me, you’ve probably bought a beaded ice cream at your local ballpark or fairgrounds during the dog days of summer. You know the kind, that ice cream that comes in tiny cryogenically frozen balls that melt as soon as you scoop a spoonful into your mouth.

My experience with beaded ice cream has been primarily with Dippin’ Dots, but today there are also a number of other beaded ice cream brands out there, the biggest non-Dots alternative being Mini-Melts.

After getting its start in Europe in the 90s, Mini Melts landed in the US a couple of decades ago when a one-time Dippin Dots distributor bought the North American rights to the beaded ice cream brand. After successfully withstanding a legal challenge from Dippin’ Dots, Mini Melts ice cream can be found nowadays across the country, sold primarily through automated kiosks that store the ice cream at negative 40 degrees Fahrenheit.

The company grew from a single kiosk in 2005 to over two thousand in the field today, but the journey to wide availability was a winding one. After experimenting with a few different kiosks in the initial years, the company started working with a single third-party vendor to build the machines from 2009 to 2019. After a decade of tinkering and adapting machines, they decided to build their own robotic kiosk starting in 2019.

Recently I caught up with Mini Melts USA CEO Dan Kilcoyn to hear the story of how the company went from a Dippin’ Dots alternative available at a few storefronts to building out a network of over two thousand automated kiosks that collectively serve beaded ice cream to the tune of 30 million servings and an estimated $150 million+ in sales per year.

So how did you first roll out Mini Melts in the US?

So we used to run physical retail locations mainly in shopping malls. In 2004, we started to get out of those retail locations, and we wanted to test the automated kiosk concept. We started with one kiosk. Unfortunately, we didn’t really know what we were doing at the time. So rather than read the manual, we essentially broke the first one, but it was a good learning experience for us, and we kind of built upon that to the kiosk that we have today.

Initially, you used a third-party kiosk?

Yes. We started with a couple of different third-party options that were out there, but some just didn’t work from a temperature perspective. We really needed to make sure that we could have our remote telemetry and we have a probe that downloads our sales information as well. So we tried a couple of options. From 2004 to 2009, it was all very experimental. Then when we were with one supplier from about 2009 to 2019 before they ultimately went out of business.

So in 2019, your third-party supplier goes out of business. Was it at that point you started developing your own kiosk?

You know, while we were taking these different third-party units, we were doing a lot of the work ourselves to make it work for Mini Melts. So prior to 2019, we started to ask, ‘What would this look like if we wanted it to be our own kiosk?’

What were the key considerations?

We wanted to know what was important to the location. Most of our locations are high-traffic tourism locations like zoos, aquariums, and theme parks. And they traditionally focus on face-to-face retail. So when we started initially with the vending machines, there was a pretty steep hill to climb because typically, if someone is at a location, they expect to be served by another human being. As we started to really hit our stride in 2019, we needed to make sure that the kiosk held the right amount of units because, in a given day, we could sell 200 or more cups of Mini Melts.

Was accommodating different forms of payment a consideration?

By 2019, we saw that our consumers really shifted from a cash-based system to credit card payment, Apple Pay and Google Wallet. So we really needed to make sure that the board on our machine was able to accept all that changing payment dynamics.

How did customers pay in the past compared to today?

In 2005, about 5% paid with credit cards. By 2019 it was 50%. Today it’s 60%. in 2019, mobile pay was less than 1%, and today it’s about 5%, growing by 2% every year.

Where is your ice cream made?

So when we started out, we started in the Philadelphia area, originally, and our manufacturing facility is in Connecticut.

And that one facility serves the entire country?

Yes. What we’ve built is we have 23 distribution centers across the US. It’s our own team that’s in 23 cities across the US. It’s our trucks, our technicians, and they’re going in and handling everything from the manufacturing of the product all the way down to the retail of the product.

How do you get the ice cream across the country?

We have negative 40-degree tractor-trailer units that we own, and we pull the tractor-trailers to our depots.

Your new kiosks remind me of a game I’d see at Chuck E. Cheese. How do they compare to the older machines?

The new ones definitely more like the claw-style game that you’re referring to. The only difference is there’s a prize every time you don’t have to worry about the robot missing. The old units are either the traditional kind of a bunker freezer where one would slide up and you would just remove a unit out, or it would work off of like a pulley system.

What is the mix of new vs old machines in the field?

About fifty-fifty.

And have you updated the old machines?

Yes. We went back and retrofitted all of the legacy kiosks in the field to make sure that they were able to take the payment systems and kind of upgrade everything from a technology standpoint.

What are some examples of locations for your kiosks?

We’re at the National Zoo in Washington DC, Philadelphia Zoo, Mystic Aquarium, Georgia Aquarium. Larger family entertainment centers would be Dave and Busters. Round One, Urban Air and Sky Zone Trampoline Parks. We’re at college campuses, a lot of rest stop areas, nontraditional retail locations.

Any one that is trending up at the moment?

Interestingly enough, our rest stop business is starting to kick up more as they add more electric vehicle chargers, because those guests tend to stay longer because they’re charging for longer. We see that there’s more of a need for retail there.

Thank you for your time.

You’re welcome.

You can watch the Mini Melts kiosks below in a video provided by the company.

The Mini Melts Automated Kiosk

I Attended a Workshop on the Impact of AI on The Food World. Here’s What We Discussed

Last month, I headed down to San Luis Obispo to participate in a National Science Foundation-funded project analyzing the impact of automation and AI on the food system. I’d been invited to participate in a workshop headed up by Patrick Lin and Ryan Jenkins, professors at Cal Poly and the project leads.

The workshop was the first for the four-year project exploring the social and ethical impacts of automation and artificial intelligence in kitchens. The project endeavors to draw out the wide-ranging implications of this technology, exploring both the impact on commercial environments like restaurants and how automation could impact the longstanding tradition of home cooking and family meals.

“This project will help to draw out the hidden and very broad impacts of technology,” said Lin at the time of the project’s announcement. “By focusing on the trend of robot kitchens that’s just emerging from under the radar, there is still time for technical and policy interventions in order to maximize benefits and minimize harms and disruptions.” 

The two-day workshop included a cross-section of academic types, chefs and food service professionals, journalists, and technology experts. It was the first of three workshops across continents to gather insights and work towards producing a report and academic curriculum centered around the intersection of food and automation and AI.

The workshop, structured as a giant whiteboard session, included expert presentations and facilitated conversations. During and after each presentation, the participants shared their thoughts on potential impacts – both direct and cascading effects – that could result from the introduction of AI in its various forms over time. While much of the conversation focused more heavily on AI in the form of automation – i.e., cooking robots – AI in other forms, such as generative AI, was also discussed.

Below are some of the key themes discussed during the two days, as well as a few of my thoughts now that I’ve had time to think through the issues since the workshop.

I’d also love to hear your thoughts on this critical topic, so please send them along!

Finally, we’ll be discussing many of these same issues at the Food AI Summit on October 25th. If this is an issue critical to you and your company, make sure to join us!

Atrophying Cooking Skills

One of the concerns raised during the workshop was the potential loss of cooking skills and culinary knowledge as we rely increasingly on automation and AI to make our meals. While it was generally recognized that robotics could take over repetitive and tedious cooking tasks, some wondered if handing over the cooking process to machines could lead to a general loss of competency in culinary arts and a homogenization of meals produced by highly automated cooking.

It’s easy to see how highly automated food prep would be extremely popular; some would hand the entire process over to the machine. However, there’s a good chance that handing off the mundane parts of cooking would give home cooks, chefs, or food workers more time to focus on creating the special touches that often make a meal great. As we have seen with the advent of digital design and art tools, there’s a possibility that those who love making food could use technology to take their work to the next level.

The Loss of Together Time

Another concern raised across the two days was the impact on shared family time by handing over meal prep and cooking to robots. Parents and other caregivers often use time in the kitchen to share lessons to help children develop motor skills, understand their heritage and develop self-confidence. Over-automation of cooking could disrupt this transfer of knowledge. Cooking has also shown many positive mental health benefits for those involved.

I think these are valid concerns, as there is a real risk of losing some of the benefits of the shared cooking process due to automation. After all, there’s no replacement for a grandchild spending time with their grandma learning how to make her special cookies and the sharing of family history that comes along with such an activity.

However, a few counterpoints. First, no one says the act of hand-making that special recipe has to be a victim of technology, and, in some ways, I think the kitchen will prove to be one of the areas where some families will insist on preserving the art and act of doing the actual cooking themselves.

And as the world becomes more digital and automated, kitchens may be a refuge for many who find the hands-on nature of making food therapeutic and fulfilling. In other words, the kitchen may be the last true ‘maker space’ left in our homes, and many will look to protect and preserve that.

Finally, average meal times shrank 5% between 2006 and 2014, a much smaller decline than we’ve seen in meal prep times as the advent of ready-to-eat meals has become more popular over the past few decades. While automation may result in faster meals, people could spend nearly as much time – or maybe more – sitting around the dinner table.

A Loss of Authenticity, Creativity, and Happy Accidents

With AI, there’s a chance recipe creation algorithms may rely too heavily on existing data patterns and therefore lack originality. There was also the concern that AI systems may limit opportunities for spontaneous creativity and the type of “happy accidents” that often lead to new recipes. One workshop participant gave an example of mistakes leading to important new dishes, like the croissant.

There was also concern that using AI to generate meal plans or recipes could result in over-standardization and homogenization, particularly if the AI systems rely too narrowly on popular recipes, which could also reduce culinary diversity.

It’s a valid concern that AI systems will generalize based on limited data sets, often creating recipes or meal plans based on popular or trending food concepts. Anyone who listens to algorithm-generated playlists by Spotify or Pandora can attest to some off-note song recommendations, and I can see how that could easily be the case with food and recipe generation. However, good technology products allow humans to reject recommendations and fine-tune algorithms, which may allow for more personalized recommendations based on a particular user’s preferences.

There’s also a real possibility that AI could lead to new and intriguing food combinations. Chef Watson and other AIs have been able to create unexpected but interesting recipes based on intelligence built into the algorithms around flavor compounds. If a restaurant or home chef can leverage heretofore inaccessible deep insights based on science and flavor research built into AI systems to create their next masterpiece, the results could be exciting.

As for the impact on cultural diversity, I think it’s important to recognize that AI systems are known to have bias problems, often hewing more closely to the worldviews of their creators and their preferred datasets. Because the world of food is one of the most important pathways for under-represented voices to connect with broader audiences, it will be critical for us to guard against the loss of accessibility and equality in the culinary world as AI and automation tools become more commonplace.

However, food AIs could be built to emphasize unique and emerging food cultures, which could be a savvy move since millennials and younger generations celebrate new food discoveries, often from cultures outside their home markets. Also, many of the creators of new food automation technology are often from markets outside our own, emphasizing food types different from our traditional fare.

This is just a few of the themes discussed during the workshop. Other themes, such as job loss and the economic impacts of automation, were also explored in detail, and I’ll have more thoughts on that later this week.

Mill Celebrates Standards Group Approval of Upcycled Kitchen Scraps As New Animal Feed Ingredient

Last week, Mill, a company that makes a kitchen scrap upcycling appliance, announced that a standards group had voted to approve a new ingredient feed definition, effectively giving a thumbs up to the output produced by the Mill appliance.

According to the announcement, the ingredient definition committee of the Association of American Feed Control Officials (AAFCO) had unanimously approved a new animal feed ingredient definition for Dried Recovered Household Food. The approval follows a recommendation from the Food and Drug Administration (FDA) earlier this year.

This is good news for Mill, which pitches its hardware and service as a way to put calories destined for landfill back into the food system through what it calls food grounds. The way Mill’s service works is the food grounds – the heated, dried-up material resulting from processing within the Mill kitchen bin – are sent back to Mill, which then turns it into chicken feed. Now, with the leading animal feed standards group giving this feed ingredient category an official thumb’s up, Mill might have just helped pioneer a new upcycling pathway for standard household food waste to make its way into animal feed.

For Mill, the news follows the recent opening of its new facility in Mukilteo, Washington, its first dedicated facility for processing food grounds into chicken feed. While the choice of a Seattle suburb might be a bit surprising for a northern California startup like Mill, it makes sense when you consider the company’s first municipal partner for its Mill service is the city of Tacoma, which is working with Mill to pilot a service which offers Mill bins and the processing service to Tacoma city residents for a monthly fee.

According to the company, while this is a significant milestone in general for this emerging category of food scraps upcycling, there are a few i’s to dot and t’s to cross.

Although the new definition still needs to clear two procedural votes later this year before its expected inclusion in the AAFCO Official Publication (OP) in January 2024, the committee vote and FDA recommendation were the most rigorous regulatory reviews required and demonstrate significant confidence in and momentum around the definition. 

With this news, I’ll be interested in watching if other consumer food waste recycling product companies attempt animal feed as a new potential service opportunity. Mill is the first company to offer an associated service with a home food scraps bin and likely has filed patents around the entire bin and service concept, but there probably will be some space for others to produce products here.

Strella Believes Its Machine Learning Tech Will Help Deliver The Perfectly Ripened Banana

Did you know that there’s a job in the banana industry called a ripener?

It makes sense, right? After all, anyone who eats bananas knows the time it takes to go from rock-hard green banana to brown mushy mess can be as short as a week. This means the banana industry has to work hard to ensure bananas ripen at the right time so they are peaking in bright, beautiful yellow by the time they show up on grocery store shelves.

Like many jobs, the ripener role relies heavily on judgment. Not that they don’t use some modern tools when monitoring and managing the ripening cycle of the banana, but from the looks of it, the ripener job seems ripe (sorry) for a Moneyball-style analytics and technology revolution.

Enter Strella. The company, which has gained traction in the apple industry for its IoT monitoring technology over the past few years, has gone bananas. According to company CEO Katherine Sizov, the company’s new AI-powered model helps them (and those working as ripeners) better decipher the signals the bananas send.

“We’ve built a machine learning model that helps us get bananas from that green to that perfectly yellow color every single time,” Sizov told The Spoon. “And the way that we do that is we measure what the bananas are telling us.”

According to Sizov, the hardware they use for banana monitoring is the same as for apples. The difference is software.

“The hardware is the same, but the algorithms are different,” Sizov said.

Sizov says that whether it’s apples or pears (fruit with longer ripening cycles) or avocadoes or bananas (fruit with shorter ripening cycles), the key indicators sending signals around the ripening stage are ethylene and CO2 emitted from the produce. The Strella hardware module has eight different sensors, sensing ethylene, CO2, and other environmental factors such as heat and moisture.

And just as with apples, the Strella technology can help determine what exactly is needed to slow down or accelerate the ripening cycle of a banana. The only difference is that things move much more quickly with bananas or avocados, which is why a job explicitly focuses on managing the process of ripening the produce.

“Unlike bananas, apples are picked perfect off the tree,” Sizov said. “And they can last a whole year in gigantic storage rooms.”

With bananas or avocadoes, the ripening process is much more closely managed. They are picked before they are ripe and then stored cold to slow the ripening until they get near the point of consumption. From there, they go into ripening rooms, and the ripener introduces ethylene gas and CO2 and adjusts the temperature to kick the ripening process into gear. And now, according to Strizov, Strella’s new banana and avocado machine-learning algorithms can help determine precisely how much of each is needed to adjust the ripening cycle to get the desired output.

Should ripeners be worried about technology taking their jobs away? Sizov doesn’t think so.

“When people are very good at their jobs, they’re always looking for tools to do better,” Sizov said. “Ripeners have a ton on their plate, they’re working 12 to 14-hour shifts, so I think they’re always looking for ways to get a little more sleep. Our tool is one way to do that.”

According to Sizov, Strella has worked with 85% of the US market for apple and pears suppliers and estimates the company has saved 20 million pounds of apples and pears from going to waste. Now, she hopes they can replicate that success in bananas and avocados.

“We’re growing pretty quickly, and we’re excited to get into bananas and avocados after having had our foray into apples for five years now.”

If you’d like to hear Katherine discuss how AI can perfect the ripening of bananas, she will be speaking at the Spoon’s Food AI Summit on October 25th in Alameda, CA! Get your early bird tickets today!

Robomart Debuts Fully Autonomous Mobile Grocery Store Concept As It Announces $2M Seed Round

Robomart, the company that introduced the store-hailing concept at CES in 2019 and started rolling out its first pilot store last year, announced a new model for its mobile store concept this week with the Robomart Haven. The Haven, which will join the first-generation Robomart (now named Oasis), will be available for retail partners starting in 2025.

Unlike the Oasis, which targets ice cream purveyors, restaurants, and cafes, the Haven resembles a convenience store on wheels. According to CEO Ali Ahmed, the Haven can stock over 300 separate SKUs and hold thousands of individual products for sale. The Haven will also act as a small walk-in mobile store, allowing customers to walk inside and shop.

And unlike the Oasis – which are retrofitted vans with drivers to pilot them – Ahmed says that the Havens will be fully autonomous. Spoon readers might remember that the Oasis (then just called Robomart) was initially intended to be a fully autonomous vehicle, but at some point, the company realized building a fully autonomous vehicle to carry its retail storefront was a lot to bite off for a first product. So instead, they chose to use drivers for their first version of the mobile store and create a fully app-controlled shopping experience (the drivers don’t interact with the customers, according to Ahmed).

But with the Haven, the plan is to make it fully autonomous, according to Ahmed.

“The Haven will not be a retrofitted van with a driver cabin,” Ahmed told The Spoon. “It will be a fully driverless offering that gives us the ability to retrofit the entire space inside to serve our customers.

Ahmed says that the company still doesn’t plan to be “an automobile company,” – meaning they still have no plans to make their own car – but instead plan to work with various automakers to convert an automobile into a Robomart Haven.

With a bigger footprint, Ahmed says future partners will be able to entirely white label and customize their Haven and will be able to have multiple types of food storage (ambient temperature, frozen, cold, or heated) within the vehicle. This differs from the Oasis, which only offers one kind of food storage (such as frozen for its ice cream partner Ben & Jerry’s).

The news of the Haven comes alongside the announcement of the company’s seed funding round of $2 million led by W Ventures with Wasabi Ventures, SOSV, HAX, and Hustle Fund, among others participating. The funding is a vote of confidence for a company that has shown significant traction over the past year, inking deals with seven partners and having a total commitment as of this week for 106 Oasis mobile storefronts.

So far, those storefronts have only been pilots, but Ahmed says they will soon be fully commercial rollouts. While he wouldn’t commit to giving a specific date, Ahmed says that the post-pilot Oasis machines will be out this year, and they will start taking orders for new ones next year. The Haven, which will roll out sometime in 2025, is also open for orders.

Barsys Makes Case For Adding Style to Bartender Robot Category With the Barsys 360

Home cocktail-making appliances have gone through lots of phases since we started writing about them in 2016. We’ve seen everything from pod-based systems from Bartesian and Drinkworks to DIY approaches like those from MrBar.io to cocktail robots with names reminiscent of 80s hip-hop artists.

And, if we’re honest, most don’t look that interesting, either presenting as something of an after-dark Keurig or a mini version of the restaurant bar dispensing system.

In other words, cocktail bots nearly always focus on utility over design.

But should that be the case? I mean, shouldn’t home appliances, especially ones focused on entertaining and leisure, actually look good? Barsys, a company that’s been making bartending appliances for the home for the past five years or so, is trying to make precisely that case with its latest product, the Barsys 360. With an interesting-looking ring-shared design allows the cocktail glass to sit within as various ingredients are dispensed from overhead, the Barsys 360 is a significant departure from any home cocktail appliance we’ve seen here at the Spoon

In fact, at first glance, it looked a little heavy on design over function, as I wasn’t sure exactly where the machine’s liquid chambers were located or how to get the liquid inside. According to the specs, it has six, and the company assured me they all sit within the 360’s ring itself. Spirits and mixers are added into the 360 via three holes at the top, using a small adapter called the “spirit funnel” seen in the rendering below. According to the company, each of the six liquid canisters can hold 900 ml in each canister (about 4 cups).

The new Barsys360 looks much different than the previous Barsys 2+, which looks like a 3D printer with a bottle-dispenser mechanism on top. The 360 also comes with a significantly lower price tag (although I’d hesitate to call the 360 cheap) at $475 for pre-orders.

With the 360 succeed? Hard to say, mainly because outside of Bartesian, the home bartender bot market has proven to be a tough market in which to gain traction. Part of the problem is most consumers have a couple of go-to cocktails they like, and, for the most part, they know how to make them. For these folks, introducing a relatively expensive machine to automate the process may seem like an unnecessary step.

However, by focusing on design and something that might look good in the kitchen or entertaining room, Barsys hopes to appeal to craft cocktail nerds who want to add a little technology-powered flair to their cocktail-making routines. And, unlike the pod-based machines, they are removing any need to rely on proprietary supplies from a startup (another big red flag for this category in the mind of consumers).

If you’re interested in a 360, Barsys is launching pre-orders this week. If you do buy one, make sure to let us know how it goes.

You can watch the hero reel video provided by the company below:

Sidewalk Delivery Startup Serve Robotics To Go Public via Reverse Merger

Serve Robotics, the sidewalk delivery robot startup that began life as a research project within Postmates called X, is going public via reverse merger, The Spoon has learned. The news, first reported in Techcrunch, marks one of the first exits for a food automation startup and the first known exit for a sidewalk delivery automation startup.

According to Techcrunch, Serve is going public via a reverse merger with blank check company Patricia Acquisition Corp. Ahead of the reverse merger, which was completed earlier this month according to filings with the SEC, raised $30 million from existing Uber, Nvidia, and Wavemaker Partners.

The news is a rare bright spot in a tough stretch for food automation startups. Companies shutting down or laying off employees has become commonplace over the past 12 months as venture capital funding dries up, with hard-tech sectors like robotics getting hit especially hard. Basil Street Pizza, Pazzi, Chowbotics, and Creator have called it quits over the past year, while others like Picnic have had to lay off employees as they struggle to raise additional capital.

The journey from a skunkworks project to a publicly traded company has been quite the journey for Serve. Initially debuted in pretty much the same form factor as today’s big-eyed delivery bot in 2018, the Serve robot has gone from a Postmates project to a division within Uber to spinout to becoming a public company in the space of five years.

Company CEO Ali Kashani has been along for the entire ride, first serving as Postmates head of special projects in 2017 when Serve was first incubated within X, then serving as head of robotics at Uber, and later becoming CEO of the Uber spinout. And as of this month, Kashani is becoming CEO of a publicly traded company delivery automation startup.

Ansā’s RF-Powered Countertop Coffee Roaster Attracts $9M in Fresh Funding

Ansā, a Tel Aviv-based builder of micro coffee roasters, has raised $9 million in funding to help the company roll out its technology into the US market, according to a release sent to The Spoon.

Unlike traditional coffee roasters, which use gas to heat the roaster chambers. The Ansā e23 micro roaster uses radio waves to heat the beans. Much like a microwave oven or newer solid-state cooking systems, this form of heating, known as dialectic heating, heats the coffee bean from core to shell.

According to the company, the system’s computer vision (provided via a built-in camera) coordinates roasting with precision application of the radio waves to transmit the energy to individual beans, creating a highly precise and homogeneously applied roast.

As automation and newer, cleaner technology enables more food processing to move closer to the point of consumption, coffee figures to be one of the leading categories in coming years. We’ve seen the success of electric-powered in-store roasters like other upstarts like Bellwether and some attempts at countertop home coffee roasters (mostly with success), but Ansā is the first to use electromagnetic heating for coffee roasting in office environments.

According to the company, they have secured commercial contracts in the US in cities such as Seattle, New York, Los Angeles and Atlanta and plan to use the funding to fund their rollout in the US.

Innit Debuts FoodLM to Power More Contextually Relevant Answers from Generative AI Platforms

Today Innit, a startup best known for its shoppable recipe and smart kitchen software solutions, announced the release of FoodLM, a software intelligence layer that helps power more contextually relevant food-related answers from generative AI large language models (LLMs).

The new platform, which itself is not a new LLM, is instead a software intelligence layer built to plug into existing LLMs to do pre and post-processing of queries to help provide better answers around a variety of food-related topics.

From the announcement:

FoodLM enables powerful semantic search for retailers to go beyond keywords and understand intent. Brands can provide consumers with highly personalized AI assistance from product selection through preparation and cooking. For health providers supporting patients with chronic diseases such as type 2 diabetes, FoodLM provides powerful science-backed assistance for healthy eating and food as medicine.

Innit CEO Kevin Brown described FoodLM as a “vertical AI” expert layer that can integrate into popular LLMs such as OpenAI’s GPT4 or Google’s PaLM. Brown compared FoodLM to what Google has done with Med-PaLM, which is Google’s medical knowledge layer that provides focused answers that are so contextually smart around medical information that it has started to pass the medical exams.

“You’re going to need the pairing of an LLM with expert training and expert systems to narrow it down for certain functions where it’s essential to be accurate,” Brown said.

The biggest concern with LLMs today is their tendency to hallucinate. Brown says that integrating with a vertical knowledge layer increases the likelihood of more relevant and accurate answers, ultimately leading to more trust in these systems.

“Food queries are one of the top use cases for LLMs, helping with tough problems like helping to manage people’s diets,” said Brown, “But only if you can trust them. If you can trust these systems and ensure they reflect key dietary and health factors, it becomes much more valuable.”

According to the company, answers are pre-processed and post-processed through FoodLM’s focused computation models, which it calls validators. The different validators within FoodLM include:

  • Nutrition & Diets: Analyzes more than 60 diets, allergies, lifestyles, and health profiles to provide detailed recommendations tailored to individual needs.
  • Health Conditions: Provides dietary guidelines, product scoring, and content specifically designed for conditions such as type 2 diabetes or hypertension.
  • Personalized Shopping: Automated grocery purchases, incorporating personalized scoring and selection of over three million grocery products worldwide.
  • Culinary & Cooking: Advanced logic to ensure that AI-generated recipes follow culinary guidelines and are cookable. Seamlessly integrates with smart kitchens, featuring automated cooking programs.

For now, Brown says FoodLM will be used by its partners through custom integrations via API. Over time, he sees the system as having a more approachable user interface where the system is used via a SaaS model.

From my perspective, FoodLM makes lots of sense for Innit. While we’ve already seen similar moves from some data-service and SaaS providers in the food space, Innit’s offering goes further and has more granular breakouts to provide specific contextualized offerings to power food-related services for their CPG, appliance, and health/wellness industries.

If you’re interested in the intersection of food and AI, make sure to check out The Spoon’s Food AI Summit, which is on October 25th in Alameda, California.

Why Big Idea Ventures’ Andrew Ive is Excited About the Japan Food Tech Ecosystem

Just over a week ago, we caught up with Big Idea Ventures’ managing partner Andrew Ive in Tokyo to talk food tech investing. Ive was in Japan for SKS Japan, The Spoon’s event that we started in 2017 in partnership with Sigmaxyz, which has grown to become Japan’s preeminent food tech summit.

While we were surprised to see Andrew in Japan, he told us it’s natural since his company has been focused on Asia since day one.

“Part of the reason why I’m in Tokyo is that we are a global firm,” Ive said. “We opened our Singapore office on the same day as our New York office because Asia was always going to be an incredibly important part of our business. And obviously, not just the business part, but the food communities.

Andrew told us he’s excited about Japan because he’s seen a real increase in activity over the past couple of years.

“I think the corporates here, they’re starting to think about how can they work with external partners, entrepreneurs, like SKS and so on, to sort of really increase the vibrancy of the Japanese food system.”

Ive also talked about Big Idea Ventures’ initiative to harness the intellectual property of universities in the US by launching companies. The fund, called Generation Food Rural Partners, announced last week that it had launched its first portfolio company, TerraSafe Materials, a material science company developing new products, coatings, and applications for sustainable packaging.

“These universities have got an enormous array of IP,” Ive said. “Traditionally, they only ever really commercialized about 1% of all of the work, which means there’s a huge treasure trove of solutions and things that can really fix a lot of the problems in the food sustainability space that they haven’t really figured out how to commercialize yet.”

You can listen to our full conversation below.

Andrew Ive of Big Idea Ventures Talks With The Spoon

A Tale of Two Ghost Kitchens: Why Wow Bao Wowed and MrBeast Bombed

This week, James Donaldson, known online as MrBeast, sued Virtual Dining Concepts, the company behind his virtual restaurant brand.

In the lawsuit, MrBeast and his legal team claim that “Virtual Dining Concepts was more focused on rapidly expanding the business as a way to pitch the virtual restaurant model to other celebrities for its own benefit, it was not focused on controlling the quality of the MrBeast Burger customer experience and products.”

The complaint goes on to say that low quality products have resulted in thousands of negative reviews and viral social media posts, including this Reddit article which showed photos of undercooked ground beef.

Above: Picture from Reddit post complaining about BeastBurger

Through its lawyers, VDC has dismissed Donaldson’s claims as “riddled with false statements and inaccuracies” and says that he is attempting to use “bullying tactics” to force VDC “to give up more of the company to him” and is using the lawsuit to “undermine the MrBeast Burger brand and terminate his existing contractual obligations without cause.”

While it’s too soon to tell how all this will shake out, there’s little doubt that the Beast Burger brand will suffer from its namesake celebrity creator publicly complaining about the quality of the food. While VDC has shown no intent to relent and shut down the BeastBurger brand, the current trajectory for the world’s most famous virtual restaurant brand doesn’t appear sustainable.

Ever since I first wrote about MrBeast’s growing disenchantment with the BeastBurger project, I started to think back to a conversation I had this spring with Wow Bao CEO Geoff Alexander. Like BeastBurger, Alexander’s company ventured into the virtual restaurant business a few years ago. However, unlike BeastBurger, there is no celebrity discord to deal with, and from the looks of it, Wow Bao’s ghost kitchen business appears to be thriving. In fact, according to Alexander, the company had just expanded its virtual restaurant footprint by over 106 restaurants in about four months, which brings the total number of virtual WowBao locations to over 700 at the time of our conversation.

So why is Wow Bao succeeding while BeastBurger struggles? From what I can tell, the two brands have three significant differences: Quality control, partner monetization, and product niche.

From a quality control perspective, Wow Bao and BeastBurger are very different. Unlike BeastBurger and lots of other virtual brands which rely heavily on its various restaurant partners to source and make the food, Wow Bao simplifies the process by delivering ready-to-steam products to the restaurants.

“We ship frozen products around the country,” Alexander told me. “If you can steam the product, you can make the product.”

That’s right; no cooking burgers, fries, or other foods, no assembling different ingredients with varying results. Hearing Alexander explain it, the Wow Bao model is the restaurant kitchen equivalent of me bringing home a bag of frozen dumplings from Costco and throwing them in my Instant Pot.

Another difference is the monetization model. According to Alexander, Wow Bao’s restaurant partners only pay Wow Bao for the cost of the food, a vastly different approach from many virtual brand management companies that take a cut of the overall revenue (while also leaving the cost of food and labor to the restaurants). After deducting labor and food, the third-party delivery fee, and a cut of the revenue to the virtual brand partner, there’s often not enough of a financial incentive for the restaurant operator (which usually has its own branded business to worry about) to give the love and attention a brand like BeastBurger needs.

The third big difference is product niche. Asian food’s popularity has skyrocketed in recent years but is still somewhat underrepresented in quick service chains compared to more standard American fare. A typical midsize suburb town in the US might have five to ten burger joints and a similar number of pizza places but may only have a couple of Asian restaurants (and often very few fast-casual or fast food variations). Wow Bao’s dumplings and buns are more likely to face less competition on third-party delivery apps than other categories.

Finally, one other difference is worth mentioning: Wow Bao is an actual restaurant chain complete with its own restaurants, while BeastBurger was born in the virtual world as a business concept, built around an online celebrity made famous not by food, but by playing video games and tracking his life via almost daily videos uploaded to YouTube. There’s something to be said for food born from an actual restaurant with an actual menu to one born out of a business plan to create a non-core business brand extension.

Beast Burger’s problems are not unique. Over the past year, it became clear that many ghost kitchen and virtual restaurant brands that rolled out in recent years would likely not survive. After Uber Eats and DoorDash began to more closely regulate and cut back on the virtual brands on their platform and chains like Wendy’s started to pare back their plans for virtual locations, it became clear the end of the wild west era in ghost kitchens was near. Now, with MrBeast’s efforts to shut down BeastBurger, we have what looks to be a definitive end to the first chapter of the ghost kitchen industry story.

The good news is some companies like Wow Bao and Hungry House are showing that there are other ways to operate ghost kitchen models and make it a win-win for both the ghost kitchens/virtual brands and their restaurant partners.

As for Wow Bao, it appears they will soon expand beyond their restaurant business and take a page out of MrBeast’s book by bringing their starting their own packaged goods business. This week, Alexander teased the release of Wow Bao retail products with a post on Linkedin.