Amazon One is expanding to its biggest area yet: the company is now testing its palm-scanning payment technology in Whole Foods, starting with a single store in Amazon’s home city of Seattle.

The company has been using Amazon One payment technology in its Amazon-branded stores in the Seattle area (including Amazon Go and Amazon Books), but the Whole Foods rollout will make the most substantial expansion of the technology yet. The company says that thousands of customers have already signed up with Amazon One.

According to an Amazon FAQ, the palm-scanning technology analyzes “the minute characteristics of your palm — both surface-area details like lines and ridges as well as subcutaneous features such as vein patterns” in order to identify a customer, allowing them to use the biometric scan as an alternative (and, theoretically, faster) method of checking out than fumbling around with a credit card or cash.

Customers will be able to register their palms at kiosks in the supported Whole Foods stores, allowing them to associate a physical credit card to that palm scan. (Amazon One users who have already registered may have to re-link their cards once to be able to use them at Whole Foods.) And of course, Amazon One users will be able to link their Prime accounts to their scans to get the subscription service’s discounts when shopping.

Amazon One will debut at the Madison Broadway Whole Foods in Seattle as an additional payment option for customers, with plans to expand it to seven other Whole Foods stores in the Seattle area over the next few months. Amazon hasn’t announced plans to further build out the palm-scanning payment system outside of the Seattle area.

All of this, of course, assumes that you’re OK with Amazon building an ever-larger database of biometric information linked to its customers, something that some experts have raised concerns about. That’s particularly true given that Amazon’s data — unlike other biometric security systems, like Apple’s Face ID — is stored in on the cloud, rather than secured locally on a specific device.