Technology
NewsOnScale Staff
October 3, 2026
Walmart has invested hundreds of millions of dollars in artificial intelligence-powered theft detection at self-checkout lanes. Cameras track items in real time. Weight sensors flag discrepancies. Alerts fire when something doesn't match.
None of it matters if the employee who responds to the alert doesn't look at the screen.
A shopper at a Florida Walmart recently watched a customer with two full carts pull baby clothes into a Walmart bag at self-checkout without scanning them. The store's AI flagged the theft in real time. An employee walked over, swiped her badge to dismiss the alert without looking at the screen showing the theft in progress, and walked away.
The customer continued scanning.
This is the gap at the center of retail AI theft detection: the technology works better than the human response to it. An alert dismissed without review is functionally the same as no alert at all.
Walmart is not alone. Target, Kroger, and other major retailers have deployed similar AI monitoring systems at self-checkout. All of them face the same problem. Self-checkout was designed to reduce labor costs. The same labor reduction that makes self-checkout profitable also means fewer trained employees available to respond to alerts effectively.
Walmart's own internal research has found that self-checkout shrink, meaning losses from theft and scanning errors, runs significantly higher than staffed checkout. Some stores have responded by pulling back on self-checkout lanes or requiring receipt checks at exits. Others have added locking cases and required staff approval for high-value items.
The retail industry's bet on AI to solve a problem created by reduced staffing has produced a predictable result: sophisticated detection paired with inadequate response. The AI finds the theft. The understaffed employee dismisses the alert. The customer walks out.
Technology doesn't solve a staffing problem. It just makes the staffing problem more visible.
NewsOnScale is owned by AMILLI AI, Corp., founded by JJ Johnson.