Rethinking Grocery: Why Smart Grocers Are “Checking Out” Autonomous Stores

A RETHINK Retail special report on the innovations in the world of frictionless shopping

Special thank you to leaders who provided RETHINK Retail with insight: Andrew Brady, Jeff Roster, Jerry Sheldon, João Diogo Falcão, Mike Fogarty, Steve Dresser, Steve Gu, Stewart Samuel and Thomas Knox.

What you’ll learn:

  • How is AI making checkout-free shopping easier, and why is it preferred by 70% of shoppers?

  • Why do we need this frictionless option, and how is it improving the grocery experience already?

  • How can mid-sized retailers afford autonomous store technology? Hint: install costs ~60% less with the latest tech.

Subscribe and Get Your Copy of This Report

This field is for validation purposes and should be left unchanged.

Before I can let you continue reading, you must first answer a riddle: Mrs. Jones walks into a store. She takes a loaf of bread from the shelf, carries it out of the store without passing through checkout, and takes it home—yet she has not shoplifted. How is this possible?

Answer: Mrs. Jones was shopping in an autonomous store. She was charged the proper amount for the bread she took, but did not need to wait in line or interact with store staff, and received a receipt just moments later on her phone.

What is an autonomous store?

Simply put, autonomous stores can run almost independently, using artificial intelligence software and other emerging technologies. As recently as five years ago, the technology needed to power an autonomous store required months of process testing and an array of costly hardware, including smart shelf sensors, smart turnstiles, and advanced high-depth cameras for computer vision.

There are several companies in this industry, but one is leading the pack by challenging the status quo to develop a solution that simplifies the process. This leader in autonomous stores is going toe-to-toe with Bezos’ behemoth—and it has a leg up. That company? AiFi, an innovative enterprise with the ambition of making AI become as accessible as Wi-Fi—which is where the name AiFi comes from—and they are starting with bringing AI to retail in a powerful way. The first to adapt a camera-only solution from a sensor-fusion model, it successfully eliminates the need for expensive weighted shelves to make product predictions.

This evolution is a huge game-changer. The technology is no longer cost prohibitive to mid-size retailers and grocers alike. With this level of innovation, frictionless checkout is becoming easier and more common for the whole world: from tiny autonomous NanoStores to larger hybrid solutions and full-scale supermarkets with the option of checkout-free capabilities.1

The growing need

Although checkout-free shopping may seem daunting at first, the world is clamoring for it, with about 95% of consumers preferring being left alone as opposed to interacting with an employee when they shop.2 For grocery stores specifically, 89% of surveyed consumers want a faster checkout, and over half believe that grocery stores need to modernize the customer experience, or they are in danger of their clientele finding food elsewhere.3 It’s clear then that it is not likely for customers to remain loyal to a brick-and-mortar location that does not attempt to adapt with the times.

Since the introduction of Amazon Go in 2018, the first truly large-scale automated store, people have had a taste and are hungry for more. Jerry Sheldon, VP of Technology at IHL Services, believes that it “wasn’t disruptive because of a volume play or that [Amazon] Go was going to take over the grocery space, but because Amazon was able to fundamentally…define in the consumer’s mind what the shopping and checkout experience could be like.” The market for checkout-free retail is projected to reach about $50 billion, with almost 70% of shoppers preferring that experience.4

Pandemic-induced progress

Of course, in the face of the COVID-19 pandemic, any technology that promotes contactless shopping is in demand. While some of the widespread secondary effects of the pandemic will fade, such as wearing masks, there are some areas that it has stamped a permanent footprint into. For example, increased remote work, online shopping and food delivery, and the accelerated movement toward reducing direct personal contact are all expected to have a permanent impact on our society, and continue moving in those directions.5

Autonomous shopping offers many health and safety advantages, from the obvious—like reducing employee-to-customer contact, as well as customer-to-customer contact while waiting in line—to more subtle benefits, such as encouraging the use of credit cards rather than unhygienic cash. Staff will have more time to stay on top of cleaning. Because of the quicker shopping experience, stores will also be less likely to reach maximum capacity, thus allowing them to serve more customers.

As Steve Dresser, Director for Grocery Insight, points out, “there are clear benefits of deploying the technology where theft and shrinkage is high,” and considering the spike in shoplifting during the pandemic,6 this technology could not have come at a better time for grocery stores.

The accelerated desire and requirement for this type of shopping is evident in Figure 1, which shows autonomous stores increasing by 50% from 2018 to 2019, to then triple the following year during the pandemic.7

That number is projected to double again in 2021, and will continue growing significantly in the years to come, though at a more manageable rate. Against the dark clouds of the pandemic, a bright silver lining shows itself as a catalyst for progress. João Diogo Falcão, AiFi’s chief technology officer, observes the push for retailers to offer more updated technology to their customers. “They know if they don’t provide that to them, their neighbor will.”8

Jeff Roster, former IHL strategist and current host of the “This Week In Innovation” podcast, admits that his own feelings on autonomous retail have changed since the beginning of the pandemic-related shutdowns. “Prior to that day I actually preferred to engage with checkout clerks,” he says, but has since come to the conclusion that since “the checkout experience has become so much less human, with plexiglass barriers and cashiers wearing masks, why not use more tech?”

Because of the rapid changes the world has had to undergo, it is a time for the clouds to make way for a new day, as we transition into a new phase of retail.

The evolution of the grocery game

A movement is taking place in the grocery world once again. Starting with the mom-and-pop stores, which were overtaken by big commerce, we saw another shift with the advent of the internet. E-commerce lets you make a wider range of purchases than were previously available, makes it easier to compare pricing, and you can do it all from the comfort of your home, or wherever you like.

The next step to revolutionize the market will be AI commerce, according to Steve Gu, AiFi’s co-founder and chief executive officer. This is partially due to the simple fact that it’s a new level of technology that makes shopping more convenient, but also because it’s ideal for smaller to mid-level stores, which are making a comeback. “They don’t have this huge overhead of massive supercenters where people aren’t shopping [as much],” Gu points out. “The mom-and-pop store that we all loved is kind of coming back, but this is only possible because of the efficiencies that can be generated through what we call AI commerce.”

So, how does it work?

Not only is the new Choice location one of the largest of its kind worldwide, it is the largest camera-only store that AiFi has launched in the U.S., as opposed to sensor fusion, which combines cameras with weighted shelves. Sensor fusion was the original design of autonomous shopping and is still being used by Amazon Go.

AiFi has engineered a way to eliminate weighted shelves from their solution without compromising accuracy, which ultimately saves businesses significant money by bypassing all the costs and maintenance that come with those sensors. This was a significant evolution of the technology made possible with data gleaned from those earlier, more complex models embraced by AiFi. The current camera-only solution relies on not only recognizing the products, but interpreting customer behavior as well, which is reinforced by merging real-world data and the advanced synthetic modeling data.

Over-instrumenting saves time and money in the long run

Because this technology is still relatively new, AiFi stresses the importance of over-instrumenting right from the earliest stages of development to help offset potential gaps in information. The more research gathered and tests run up front, the easier it is for the AI to learn to operate on its own without needing all the extra equipment anymore. It’s the same as with human intelligence, evolving from infancy’s complete dependence on the parent into someone that can make decisions on their own.

This is why the combination of real-world data and simulated data is so essential to AiFi’s success. Although a unique AI “brain” is created specifically for each store and has a large amount of information ready to go, it continues to learn more about customer behavior and product variance every day. Gu points out that since there is a limited number of transactions per day, their simulation engine serves to optimize how much the AI can learn in a shorter period of time, so it continuously becomes more and more accurate.

Making AI technology more affordable and accessible

The incredible complexity of this autonomous solution makes it seem unattainable for those non-multi-billion-dollar companies, but AiFi has found solutions to make it affordable from multiple angles. Simply by transitioning into camera-only systems, installation costs were reduced by 60%, according to Falcão.

In addition, the efficiency of AiFi’s AI results in uniquely low compute requirements, so they are able to utilize 5G, allowing the stores to have more mobile coverage without having to worry about maintaining hardware themselves. By removing the need for fiber connections, autonomous stores can now open anywhere they like, and existing stores can incorporate the technology without spending thousands on fiber connections.

Verizon’s 5G Ultra Wideband network was able to make this possible, most recently at the 105th running of the Indianapolis 500.9 The shipping container-sized NanoStore allowed customers to quickly grab a beverage or snack and get back to their seats without missing what they came to see.

Vice president of sales for Verizon, Andrew Brady, comments that “Giving race fans the ability to experience the power of Verizon 5G through AiFi’s autonomous shopping is an exciting moment for us,” and adds that their partnership with Penske and AiFi holds a passion “filled with industry-driving innovation.”

Effects on labor

The term “cashierless” may instill fear in some, implanting thoughts that jobs will be eliminated left and right. As retail workers make up 6.3% of the U.S. labor force,10 this is a legitimate concern for the economy as a whole. However, there are several factors to consider on this subject:

  • The various forms that “cashierless” stores can take, not all of which are unstaffed
  • The fact that AI is poised to replace a large portion of tasks, not necessarily whole jobs
  • The opportunities that will be opened up once employees are freed from these tasks

True, AiFi has deployed 400 square foot NanoStores, which are essentially walk-in vending machines that operate 24/7 and only need staff for stocking and cleaning when necessary. That is not their only option, though, as they also offer larger cashierless stores and are even beginning to offer hybrid solutions with both cashier checkout and frictionless choices, which Falcão says are at once “the most complex, as well as the most in-demand.” Providing both options allows customers to choose their experience and transition more gradually from the traditional setup into the automated version. Steve Gu, co-founder and CEO of AiFi, says this helps to introduce customers to “the pleasures of no more frustrating waiting in line while simultaneously avoiding crowds, which is especially important during the pandemic.”

Even the fully autonomous grocery stores will still be staffed, but the employees will have more time to stay on top of stocking and answer customer questions without having to worry about jumping on a register at any moment. Thus, rather than disposing of jobs, the work that retail employees do will be more productive and purposeful. According to the McKinsey Global Institute, current technology has the capacity to automate approximately half of retail employees’ duties, but only about 5% of entire jobs can be fully automated.11 This means that companies could theoretically spend the same on labor while doubling productivity and providing more interesting jobs with useful skills.

Bringing back customer loyalty

With so many stores—big, small, online, in-person—customers are stretched thinner than back in the days of “the corner drugstore.” Loyalty programs strive to close this gap to some extent, but there is more we can do.

Personalized shopping

Although people are often unnerved by the accuracy of targeted ads through the use of data tracking, it is undeniably convenient. Gone are the days of sitting through ten commercials on television, only for about four of them to apply to you and maybe one of those piquing your interest. The internet has a profile for everyone and tailors each person’s advertisements to what they might need, or search for, or didn’t realize they wanted till they saw it.

Unfortunately, personalized shopping is difficult to translate back into brick and mortar stores—but not impossible. Not only is it possible, but it’s very close on the horizon. “We can take all of the data that you’ve accumulated over the years through online shopping, and leverage this within the store, and vice versa,” says Thomas Knox, AiFi VP of Product. “From the retailer’s perspective, they’re able to push targeted campaigns, driving product purchase conversion, and you as the shopper get access to discounts, promotions, and other interesting actions on items that you may have been interested in.”

With this technology, even the displayed price can be automatically adjusted depending on customer proximity to encourage store loyalty. So for example, a customer with a loyalty card may receive a lower price on a product than a customer without one, and the shelf’s digital price label will alter to reflect that.

To conclude

With this groundbreaking technology, you don’t have to be a global superpower to offer your customers the modern solutions they’re looking for. Samuel points out that it will promote growth on more than just the customer side, saying “Generally, the technology supports wider digital transformations that position retailers as cutting edge, making them more attractive to investors.”

Whether you’re an international giant or a modest mid-sized store wanting to grow, AiFi’s platform, OASIS, can be customized to fit those needs. As Falcão illustrates, stores can decide whether they want to use “their application or their discount cards, take credit cards or not take credit cards. All this customization allows the retailer to shape the shopping experience to fit their brand.” They have the technology, and you plug in the specifications that will help your store thrive.

Autonomous stores are no longer an ethereal dream for the future, but a present and advancing reality that is becoming available to a larger scale of retailers. The technology is here, and the most innovative companies are taking action. In fact, AiFi’s growing waiting list is ready for you whenever you want to outpace your competition in the race toward becoming the next store of tomorrow.


  1. AiFi – About AiFi | Future of Retail l Retail Innovation
  2. Retailers, 95% Of Shoppers Want To Be Left Alone In-Store (
  3. Survey: Grocery stores need to ‘enter the modern age’ (
  4. Retailers go cashierless to improve customer experience | APEXX Global
  5. All change: how COVID-19 is transforming consumer behaviour | Swiss Re
  6. Shoplifting has increased during pandemic, experts say – The Washington Post
  7. Autonomous checkouts: number of stores worldwide 2018-2024 | Statista
  8. RETHINK Retail Podcast – Retail Rundown: Outlook for Summer 2021 – with João Diago Falcão and Jeff Roster
  9. AiFi, Verizon Accelerate Autonomous Shopping at the Indy 500 (
  10. A Profile of the Retail Workforce (
  11. The future of work in retail automation | McKinsey

Go to Top