Walmart Inc. is test-driving artificial intelligence technology at one of its busiest Neighborhood Markets in the U.S., its Levittown, N.Y., location, and calling the experiment, Intelligent Retail Lab or IRL.
The technology was created and designed for the store by Walmart’s ring-fenced incubator, Store No. 8.
The retail behemoth said the application of AI in e-commerce has become de rigueur, but the technology hasn’t made its way to brick-and-mortar en masse.
Walmart is applying AI to inventory to improve the customer experience and satisfaction, using techniques such as training AI models with synthetic data, which can help associates keep better track of inventory and know when items need to be restocked.
AI can ensure that online orders are always ready for in-store pickup and make it easier for customers to return online purchases to the store, creating a more seamless experience. It can eliminate brick-and-mortar pain points such as having to hunt for a shopping cart, and long lines at the cash register by monitoring the situations.
The technology can also improve the safety and cleanliness of stores, alerting associates of spills and other hazards.
IRL’s AI-enabled cameras and massive data center can provide real-time, precise information to associates about the need to restock items, or relay via app when to remove from shelves product that’s not fresh, such as produce or meat.
Mike Hanrahan, chief executive officer of IRL, said of the 50,000-square-foot Levittown location, “It’s a real store with real customers and real sales and if we can get the technology working in this store, we know that it can have broad applications across the Walmart chain.”
Hanrahan stressed that Walmart’s exploration of AI will stay grounded and focused on practical uses and solutions that can benefit consumers today, rather than rushing into futuristic concepts that will only be feasible years from now.
“IRL will be in data-gathering mode in its early days,” Hanhahan said. “The focus will be on learning from the technology and not implementing changes to operations in haste. You can’t be overly enamored with the shiny object element of AI. There are a lot of shiny objects out there doing things we think are unrealistic to scale and probably, long-term, will not beneficial for the consumer.”
IRL uses a combination of cameras and real-time analytics to automatically trigger notifications to internal apps used by associates. Information is gathered throughout the store using sensors, cameras and processors with the hardware connected by enough cabling to climb Mt. Everest five times and the processing power to download three years’ worth of music, the retailer said.
Walmart debated about how much of the technology it should make visible to customers, and decided that exposing it would help shoppers get a better understanding of how AI works. The store also has multiple information stations that explain in some detail various aspects of AI.
Servers are housed in a glass-encased data center that’s bathed in blue light. Displays, including an interactive wall, demonstrates how AI can estimate body positioning. There’s a welcome center that offers a deeper dive into technology specifications and answers common questions.
While AI isn’t new — it’s been used in online search results for years — Walmart said it wants to understand in real time, its potential applications for stores. So far, at IRL, AI is freeing associates up and allowing them to engage more with the customers, the retailer said.
And in an example of its high-low functionality, Kering’s chief executive officer François-Henri Pinault touted its use of AI at its annual meeting Wednesday.