Where this conversation describes AI being used to build or deliver a product. Part of the AI in practice tracker.
Logistics & fulfilment
StockX runs a machine learning model on order characteristics and seller behavior before packages arrive to separate them into verification lanes by risk and experience level.
“before that truck even pulls up, one of the things we're going to do is run every single order through a machine learning model that we have. And that's going to look at order characteristics, seller behavior, different product risk factors, whether we've seen verification issues with products to begin with.”
Quality & testLumafield CT scanner
StockX uses a Lumafield CT scanner on products over $3,000 to image internal structure, identifying air pockets, component bonding, and speaker placement to distinguish genuine items from sophisticated counterfeits that look correct externally.
“We have a Luma field CT scanner, we'll image the product, and that can give us better insights into just internally, how it's structured, and whether we think that those characteristics map to the characteristics of a real product.”
Quality & testLumafield CT scanner
The CT scanner reveals internal component placement in counterfeit AirPods, such as misplaced speakers and wires, which external inspection cannot detect, allowing StockX to identify fakes that can still pair with iPhones.
“as soon as you sort of scan that item, you're going to be able to see, hey, the speakers aren't quite where they ought to be some of these wires here that I see on the, you know, the bona fide product, they're out of place.”
Engineering & simulationCT scanner
StockX CT scans new releases to understand the internal structure of genuine products, creating knowledge guides and quizzes for verification experts to study before manual inspection begins.
“when new products are released, we're going to be, um, getting our hands on those products, CT scanning those products, RFID scanning those products and doing everything that we can to understand what is what should a real product look like?”