Close Ad

The Spoon

Daily news and analysis about the food tech revolution

AI

Join Us Today as We Discuss How Artificial Intelligence Will Impact Culinary Creation

Admit it: you’ve probably played around with making recipes using AI. At this point, most of us have.

If you’re like me, the early results were… rough. But over time, general-purpose LLMs have become surprisingly good at whipping up recipes. Still, there’s a long way to go before AI becomes a true sous chef in our kitchens, and plenty of questions remain about where this is all going.

To help us explore what’s next in this month’s edition of our Food AI Co-Lab, we’re joined by two people who’ve been working at the intersection of AI and cooking for nearly a decade: James Briscione and Lav Varshney, co-creators of Chef Watson—the world’s first culinary AI. Their latest project, CulinAI, is an AI-powered app designed to create personalized meal plans.

Want to join the conversation, ask questions, and see where AI cooking is headed? Register for today’s Food AI Co-Lab here.

ClearCOGS Raises $3.8M its AI-Powered Forecasting Software That Helps Restaurants Reduce Waste

AI-powered restaurant forecasting startup ClearCOGS has raised $3.8 million in an oversubscribed seed round led by Closed Loop Partners, with participation from Myriad Venture Partners and Hearst’s Level Up Ventures. The funding includes $2.3 million in new capital and the conversion of $1.4 million in pre-seed investments. The company’s software provides predictive analytics to assist operators in making decisions around food prep, ordering, and staffing, with an emphasis on reducing food waste and improving operational efficiency.

Company CEO Matt Wampler told The Spoon that he came up with the idea of ClearCOGS during the pandemic. He’d been being laid off and was exploring coding and analytics, when he discovered his cousin who ran a Jimmy John’s franchise was still using a decade-old Excel forecast. Wampler wondered if AI could help create a better predicitve forecasting tool, and before long he had teamed up with Osa Osarenkhoe to build a solution that uses machine learning and time-series forecasting that currover 100 million data points a day.

When ClearCOGS participated in our first virtual Food AI Summit a couple of years ago, Osa and Matt had started experimenting with leveraging large language models (LLMs) like those from OpenAI to create an interface for their forecasting tool. I asked Matt how those experiments with LLMs had gone.

“We did a whole big thing with it… It didn’t go well,” said Wampler. “And it wasn’t from a technical standpoint. It was from the standpoint of the restaurant brands we were talking to… they were like, ‘Look, my general manager can either just get on and play with your AI bot and it’ll tell them, or you can just send it to them? Just send it to them.’”

Wampler said the LLM interface wasn’t the problem. It was just that operators didn’t want to interact with it at all. Instead, they just wanted the answers delivered to them, simply and directly, through email and integrations with solutions from Toast or SevenRooms. This experience reaffirmed Wampler’s belief that proprietary forecasting (and not LLM-powered conversational AI) is where ClearCOGS can deliver the most value.

“LLMs are kind of a commodity at this point. Proprietary data sets are what really matters… You still have to be able to provide a fundamental business value before that AI is really helpful.”

While many platforms offer dashboards or raw analytics, ClearCOGS focuses on delivering direct, decision-ready insights to restaurant managers. This is central to how he differentiates the company:

“If you’re a brand, you probably have 20 or 30 questions that you have to answer every day… We go really deep on those and provide a systematic way of delivering those to your operators every day.”

With the new capital, ClearCOGS plans to accelerate its product development and customer acquisition efforts, with an emphasis of better positioning itself in the food service sector. The company currently serves a customer base of 100 brands in four countries, and Matt says they plan to continue building a lean team, prioritizing automation and AI over headcount.

Food Assembly Robot Startup Chef Robotics Raises $43M Series A as it Reaches 40 Million Meal Milestone

Chef Robotics has raised $43.1 million in a Series A round to accelerate deployment of its AI-powered food assembly robots, the company announced today. The funding brings the San Francisco-based startup’s total capital to over $65 million, including equipment financing. Avataar led the round, with participation from Construct Capital, Bloomberg Beta, and others.

Founded in 2019, Chef Robotics is building what founder and CEO Rajat Bhageria calls an “AI platform for food.” Rather than building a single-purpose robot, Chef’s system is designed to work in diverse food production environments—learning and adapting through software to new tasks like portioning, topping, or filling.

When I first got a peek at Chef’s system last year, I was intrigued because the company had struck a balance that seemed to elude many food robotics startups. While startups in this space seemed to make either high-volume solutions with limited customizability or use off-the-shelf robotic arms that aren’t made for true high-production, Chef has built a flexible and scalable robotics platform that can be customized for any number of high-volume food production environments.

That’s because while many robotics companies focus primarily on hardware, Chef’s approach centers on a software layer that enables “Embodied AI”—giving physical robots the intelligence to operate autonomously in real-world conditions. Chef’s system combines a robotic arm with AI models trained on millions of real-world examples. These models, powered by production data from early customers like Amy’s Kitchen and Fresh Prep, allow the robots to generalize across new ingredients and dishes. To date, Chef Robotics has helped assemble over 40 million meals.

From the company’s announcement: When we thought about starting with restaurants, we ran into the chicken and egg problem – to enable robots that are flexible enough to add value, we need a highly capable AI, but to get a highly capable AI, we need real-world training data from the customer sites…. Thus, we decided to initially deploy robots in high-mix (read as highly flexible) food production and manufacturing environments where Chef could partially automate a food operation and thus add value in production to customers without requiring 100% full autonomy from the get-go. We built Chef’s systems on modern advancements in AI to make them highly flexible and adaptable enough to “pick” and plate almost any ingredient, no matter how it’s cut, cooked, or grown; this makes them an ideal solution for assembling or plating food.  

The new capital will support scaling up deployments and building out Chef’s sales and marketing teams. The company is currently active in the U.S. and Canada, with plans to expand into the UK next year.

The Food & Retail AI Rollup Continues as Crisp Buys Shelf Engine

Crisp, a New York-based retail data company, has acquired Seattle-based startup Shelf Engine. Founded in 2016 by Stefan Kalb and Bede Jordan, Shelf Engine specializes in using machine learning to optimize ordering processes for perishable goods, with the goal of reducing costs and minimizing food waste. The platform is now in use across more than 7,000 stores. Terms of the acquisition were not disclosed.

Shelf Engine was one of the earliest adopters of AI technology in food retail aimed at reducing waste and optimizing fresh food ordering. Kalb, who launched a food distribution business at 23 and holds degrees in actuarial science and economics, developed the idea during a 2014 ski trip with his friend and engineer Bede Jordan, a former Microsoft HoloLens developer. The pair questioned why food industry processes and systems remained outdated:

“Could we create a platform that enables retailers to buy food and eliminate significant waste? Could we create a platform that eliminates redundant busywork between vendors and retailers? Could we create a more perfect marketplace?”

These questions led Kalb and Jordan to develop a product designed to drive the food industry towards greater efficiency through technology.

Kalb reflected on the deal on LinkedIn:

“What started as a side project with my good friend Bede turned into a platform that’s now in over 7,000 stores across the U.S., helping reduce millions of pounds of food waste. It’s been eight years of wild highs, humbling challenges, and so much growth.”

Crisp plans to integrate Shelf Engine’s advanced algorithms into its commerce platform. The company believes the integration will help its retailer customers optimize in-stock inventory, improve shelf management, and drive revenue in an increasingly margin-sensitive retail environment.​

“Joining forces with Crisp allows us to scale our proven technology and deliver greater value to retailers and their supplier partners,” said Kalb. “Together, we will set a new standard in forecasting and inventory management, helping our customers thrive even in challenging market conditions.”​

This news is yet another in a series of acquisition announcements for early pioneers who are building technology leveraging AI to optimize different parts of the food value chain. Earlier this month AI Palette was gobbled up by trend forecaster Global Data, and before that Spoonshot was acquired by Target. Like other buyers in these deals, Crisp provides predictive intelligence software and services and is buying Shelf Engine to improve their AI insights capabilities.

Unlike these previous deals, Crisp and Shelf Engine focus more on retail and supply chain commerce optimization, which is one of the areas that is seeing the greatest leaps forward in productivity and cost-reduction. My guess is Afresh, which is similar to Shelf Engine and remains independent following its $115 million in series B funding in 2022, may also be one of the next companies gobbled up as bigger software and supply chain players look to add AI capabilities to their products.

Bridge Appliances Deploys Egg-Making Robot at First Customer

Five years ago, the cofounders behind Bridge Appliances stood in line at a busy breakfast cafe. As minutes ticked by, frustration turned into inspiration. They wondered: What if the preparation of eggs could be automated? That simple question led to the creation of OMM, a countertop egg-making robot. Now, half a decade later, the Bridge team is back in a coffee shop, deploying their robot for their very first commercial customer, Beantrust Coffeebar in Beverly, Massachusetts.

Bridge cofounder Connor White recently described spending the past two months embedded at Beantrust, collaborating closely with owner Erik Modahl and his team. According to White, working alongside baristas, listening to customers, and absorbing the café’s unique culture and operational flow allowed Bridge to tailor OMM’s integration precisely to Beantrust’s specific needs.

As I wrote last year, OMM cooks two eggs in roughly two minutes, enabling Beantrust to serve around 60 eggs an hour. White notes they’ve already seen promising results, with the new sandwich lineup boosting average ticket values by 15%. Currently, one in five customers chooses to add a freshly made sandwich to their coffee order (and that number continues to climb).

This marks a significant milestone for Bridge, which raised $2 million in seed funding from Steve Papa, one of Toast’s earliest investors, in 2021. Moving forward, Bridge is likely to see more growth among small coffee shops or similar establishments that lack full kitchens or grill cooks but still wish to offer breakfast. However, they will need to raise considerably more funding to scale effectively, or they could be a potential attractive acquisition candidate for a company such as Middleby.

The Spoon Discusses The Current State of AI Workflows With AI for Humans’ Gavin Purcell

In this latest edition of the Spoon Full of AI Podcast – the podcast where I talk to leaders who are using AI to transform their business (in food and beyond), I catch up with AI for Humans host Gavin Purcell to discuss the rapid advancements and complexities emerging in artificial intelligence. We talk about Google’s new Studio AI platform and its potential as an all-in-one solution for content creation, and we both have reservations about Google’s use of an older image generation model (Gemini 2), highlighting its limitations compared to newer versions like Gemini 3.

AI Studios? with Gavin Purcell

Gavin talked about the transformative shift from traditional file-based computing to generative AI, as noted in Nvidia CEO Jensen Huang’s recent keynote. He make a great point talking about how this will be difficult for mainstream adoption, where users expect simplicity akin to social media sharing, yet AI tools remain complex and imperfect.

We both talk about our frustrations with current AI workflows, where we compared the cumbersome nature of local models like Comfy UI to more streamlined platforms such as Ideogram and MiniMax. In the end, we both want a unified, user-friendly AI platform but recognize that technical and creative challenges remain significant hurdles.

We also discuss who we predict will be winners in the AI platform space (no spoilers!), so make sure you listen to see who we pick!

You can watch the full video above or click play below, or find this podcast in Apple Podcasts, Spotify or wherever you find your podcasts.

Marine Biologics Debuts SuperCrudes, the ‘World’s First Programmable Biomass’

This week at Future Food Tech, Marine Biologics will showcase its technology—a platform that digitally maps the biochemistry of seaweed and its properties – which the company claims has created what is the world’s first programmable biomass.

According to the company, its proprietary cheminformatics platform, MacroLink, maps the molecular compositions of seaweed and stabilizes them into customizable, liquid-based macroalgae solutions called SuperCrudes. SuperCrudes, which are liquefied blends derived from seaweed harvested from specific geographic locations, are analyzed to document their mineral, protein, and carbohydrate compositions. This process allows Marine Biologics to benchmark each product similarly to how oil refineries grade petroleum based on extraction origin.

The company was founded by former crypto industry executive Patrick Griffin, who told Wired he transitioned from crypto to building a platform that could digitally understand seaweed after a surfing accident caused him to reevaluate his priorities.

Ironically, the accident reignited his enthusiasm for the ocean. Griffin recognized a gap in the climate resiliency market concerning fundamental building materials. Even if all other global climate resiliency efforts—such as vehicle electrification and renewable energy investments—were successful, products would still be largely built upon plastics or other petroleum-based materials.

“The chemicals and materials that we use today are, by and large, built on petroleum,” Griffin says. “It’s the last piece of the puzzle you’ll really have to chip away at to make a significant impact.”

Griffin believes his company can substantially reduce the costs and variability traditionally associated with bio-based sourcing by standardizing seaweed extracts into reliable and predictable ingredients.

The broader macro-trend of computational biology is one we’ve been following closely here at The Spoon. Examples include using AI to accelerate pathways toward gene-edited seafood or tracking startups like Shiru, which aim to build food input discovery engines by mapping food’s biological building blocks. Marine Biologics’ technology is particularly intriguing because it closely connects to a specific natural biomass that has previously struggled to gain traction in consumer products—both food and otherwise—due to limited understanding and internal expertise within larger CPG brands.

With this coming-out party for its Macrolink platform, the California-based startup is hoping to change all that, at the same time becoming both a formulation platform as well as a provider of inputs (SuperCrudes) to CPG brands looking for more sustainable inputs.

AI-Powered CPG Trend Forecaster AI Palette Gets Gobbled Up

AI Palette, a company that uses AI to help CPG brands anticipate consumer tastes ahead of market trends and better predict new product success throughout their lifecycle, announced today it was acquired by market research company GlobalData.

AI Palette launched within the last decade alongside a cohort of startups—including Analytical Flavor Systems, Tastewise, and Spoonshot—that began utilizing machine learning and big data analysis tools. These tools enabled CPG brands to move away from traditional product ideation and surveys, uncovering hidden insights more rapidly through AI. Like many similar startups, AI Palette started talking up their generative AI bona fides over the past couple of years.

“The integration of AI Palette supercharges our ability to help CPG brands innovate smarter and faster,” said Mike Danson, CEO of GlobalData Plc. “Together, we are setting a new standard for AI-driven intelligence in the consumer space.”

In some ways, this deal resembles Spoonshot’s acquisition a little over a year ago, when legacy market intelligence provider Target Research Group acquired the AI startup. GlobalData Plc, which places greater emphasis on pure data analytics than Target, is a company that has grown primarily through acquisition. From what I can see, AI Palette represents GlobalData’s first pure-play AI company deal.

With most major food-brand CTOs currently exploring—or being directed—to leverage AI, it seems likely that acquisitions of vertically specialized platforms will continue over the next 12-24 months. Consultancies and market intelligence providers will probably lead this trend as incorporating AI into intelligence toolsets becomes essential to maintaining relevance..

McDonald’s is Creating Virtual ‘AI Managers’ for Its Restaurants

Want some AI with your Big Mac? McDonald’s is about to serve it up in a big way.

According to a story in the Wall Street Journal today, McDonald’s is undergoing a technology overhaul across its 43,000 restaurants, implementing internet-connected kitchen equipment, AI-driven drive-throughs, and tools for managers. McDonald’s is partnering with Google Cloud to deploy edge computing technology, allowing restaurants to analyze data locally rather than sending it to the cloud. This setup helps predict equipment failures—such as fryers or ice cream machines—before they occur and ensures order accuracy through AI-powered cameras.

We’ve written quite a bit at The Spoon about McDonald’s experimentation with AI at the drive-thru, but this story shows they are looking to use generative AI for customer interaction and beyond, including exploring generative AI virtual assistants to handle managerial tasks.

From the WSJ: Edge computing will also help McDonald’s restaurant managers oversee their in-store operations. The burger giant is looking to create a “generative AI virtual manager,” Rice said, which handles administrative tasks such as shift scheduling on managers’ behalf. Fast-food giant Yum Brands’ Pizza Hut and Taco Bell have explored similar capabilities.

This story comes just a day after Taco Bell talked up their own AI initiatives, including their ‘Byte by Yum’ AI tool designed to assist restaurant managers with tasks such as labor and inventory management. The AI can manage schedules, assist with drive-through orders, and suggest operational changes based on competitor activity, aiming to optimize employee efficiency without reducing labor costs.

Join Us Today as We Explore Future Scenarios of an AI-Powered Food System

Ever think about all the different scenarios that AI could ultimately unleash on our food system?

Us too, so we decided to invite noted food futurist Mike Lee to talk about it on our latest edition of the Food AI Co-Lab.

Join us today at 8:30 Pacific to explore the different scenarios that AI could unleash upon us.

In this session, we’ll discuss

  • AI & Regenerative Food Systems – How can technology help restore biodiversity, improve soil health, and create more resilient agricultural ecosystems?
  • Personalized Nutrition & Food Sovereignty – Can AI make food systems more inclusive, culturally relevant, and tailored to individual health needs while ensuring accessibility for all?
  • Circular Economy – How can AI-driven solutions reduce food waste, optimize supply chains, and create more efficient, closed-loop food systems?
  • Ethics & AI in Food – As technology advances, how do we ensure that food innovation remains fair, transparent, and truly benefits people and the planet?

You can watch it below or join us on our interactive Livestream to ask Mike questions!

Exploring Future Scenarios of an AI-Powered Food System

Fast-Growing Restaurant Chain MOTO Pizza Is Building an End-to-End Pizza Robot

Last month in Las Vegas, we sat down with one of our favorite restaurateurs, Lee Kindell of MOTO Pizza, to discuss his vision for the future of restaurants. Lee was in town to speak onstage at CES about the industry’s direction, so we made sure to catch up with him for a one-on-one interview.

One revelation that surprised us during our conversation was that MOTO is developing an end-to-end pizza robot—one that automates the entire process, from dough preparation to boxing the finished pizza.

“Our robot is going to be fully autonomous,” Kindell explained. “It takes the pizza from refrigeration, brings it out, proofs it, tops it, cooks it, finishes it, cuts it, and boxes it. So that’s what we’re building right now. It’s truly end-to-end, and that’s what excites me the most.”

As someone who loves both pizza and robotics, I was intrigued by MOTO’s move to develop its own technology. To clarify, I asked Kindell directly about his plans. He confirmed that MOTO is indeed building a fully automated pizza-making solution—one that could potentially integrate with existing automation partners, such as Picnic (Picnic’s pizza robot adds sauce, cheese, and toppings but doesn’t handle cooking, cutting, or boxing).

MOTO has been expanding rapidly, entering new cities and sports venues. Kindell, who started as a hands-on pizzaiolo mixing dough by hand, became a firm believer in automation after an arm injury forced him to adopt a mixer. That moment reshaped his perspective—he realized that automation wasn’t just about efficiency; it was a tool to scale his business while maintaining quality.

Now, Kindell and MOTO are taking that mindset a step further, developing an end-to-end pizza robot to help the fast-growing chain keep up with demand and reach more customers than ever before.

You can watch our full conversation below.

Moto's Lee Kindell on Using AI & Robotics to Make Pizza

While Amazon Struggles With Futuristic Retail, Sam’s Club is Doubling Down on Computer-Vision To Aid Store Exits

An interesting set of stories came out over the past week, which showed how two retail giants are leveraging technology to help consumers get out of the store faster.

One one hand, you have Amazon halving its Go store count to 16 locations, shifting focus to licensing its “Just Walk Out” technology to third-party retailers.

On the flip side, Walmart’s bulk membership club Sam’s Club is investing more heavily to use computer vision to aid shoppers to get out of the store faster.

The key difference? Sam’s Club is using computer vision for receipt verification. Previously, members had to stop at the exit while associates manually checked receipts, causing bottlenecks. Now, cameras at the exit automatically scan carts and verify purchases, allowing members to walk out without interruption. AI works in the background to refine accuracy, while employees are freed up to assist shoppers rather than policing receipts.

Contrast this with Just Walk Out, which is Amazon’s effort to eliminate the checkout experience all together which, it appears for many people, is still too weird and feels a little to close to shoplifting.

As I wrote last year, Just Walk Out is “a radically tech-forward evolution of checkout, but one in which Amazon appears to have widely overestimated just how many people would use it and how easy it would be to implement. Self-checkout fits most shoppers’ needs when they are in a hurry, and there aren’t that many situations where consumers feel they need to skip checkout altogether.”