A store with no checkout lines? According to Amazon, it’s right around the corner, at least in Seattle.
According to the online giant, Amazon Go is a brick-and-mortar store in which customers can walk out of the store without a check-out register and have the items automatically billed to their Amazon accounts.
Using the Amazon Go app, the customer enters the store, takes the products they want and then walks out when finished.
The app keeps track of what is taken, so you don’t have to spend time scanning them or manually entering them in.
Using what they say is the same technologies used in self-driving cars, their “Just Walk Out” technology:
Automatically detects when products are taken from or returned to the shelves and keeps track of them in a virtual cart. When you’re done shopping, you can just leave the store. Shortly after, we’ll charge your Amazon account and send you a receipt.
The store at 2131 Seventh Avenue, on the corner of Seventh Avenue and Blanchard Street in downtown Seattle is currently only available to Amazon employees while it’s being tested, but will open to the public in early 2017.
One employee who bought items in the store on Monday, called the experience “empowering” for customers.
“I got some lunch, some cookies and some cereal,” he said. “It’s really convenient for these urban settings. You can just walk in, pick a few things up and walk out.”
He told KIRO 7 he saved what was likely 15 minutes that he would have spent in a traditional store, waiting to pay.
Amazon declined to do any interviews about the store or the technology on Monday. So KIRO 7 took the company’s promotional video to Dr. Pedro Domingos, a University of Washington professor of computer science and engineering.
Domingos studies the same type of technology in the Amazon Go store.
“What’s happening right now, right – you don’t see it in this video. But Amazon has cameras. It has video cameras in the store,” Domingos said.
He said ‘machine learning’ is when computers learn to do things from data, rather than having to be programmed by humans.
So the system inside the Amazon Go store is learning what people are buying by watching them through the cameras, and following them around the store.
The system has learned, for example, that when one puts an item in a bag, that item has not disappeared. It has simply been hidden from view. Domingos said it’s the same concept as teaching a baby how to play ‘peek-a-boo.’
“You know when they’re very fascinated by peek-a-boo? This system has mastered peek-a-boo,” he said.
He said the system is also able to watch facial expressions, to see what a person might be thinking when considering whether to buy a product. That would help with data on how to stock items and market them to consumers.
He said shoplifting in this setting would become almost non-existent. In order to come into the store, one must be identified through the app on the phone. So Amazon knows who each person is and would charge the person accordingly for anything taken out of the store.
KIRO 7 also asked Domingos what else this system can learn about customers.
“They can learn all sorts of things,” Domingos said. “Her facial expressions as she’s looking at the shelves, right? There’s computer vision, machine-learning algorithms that can understand what emotions you’re feeling from your facial expressions. They’ll actually be able to look at her and like, what she’s thinking when she’s looking at one of the cupcakes. You know, does she scrunch up her face?”
That type of machine learning can then inform Amazon on how to stock the shelves and how to market items to people both in the store and online.
Domingos said he believes prices in these stores would be cheaper, since there would be less labor cost without people checking out customers.
The store, which is about 1,800 square feet, is compact so customers can get in and out quickly.
It offers ready-to-eat foods, grocery staples and meal kits that can be made at home.
Customers will need an Amazon account, a supported smartphone and the Amazon Go app.
For more information, watch the video below or visit http://amazon.com/go.