Scandit has launched “Self-Checkout Loss Prevention,” its new real-time vision AI solution that helps retailers detect loss patterns, whether accidental or intentional, as they occur. The solution enables customers to correct their mistakes autonomously, without staff intervention in most cases, to keep the checkout flow uninterrupted.
Scandit’s entry into the loss prevention market addresses one of the retail sector’s most visible tensions: reducing losses at self-checkout kiosks without creating friction that frustrates customers or overburdens store teams. Currently, self-checkout manages 54% of transactions in stores that offer them. These stores report losses 33% higher than similar stores that do not have self-checkout, and losses rise by an average of 22% in the year following installation, according to the ECR Retail Loss 2026 Self-Checkout Loss report. According to a European retailer, random checks can last more than two minutes, causing friction with customers and consuming staff time.
How the solution works
The “Self-Checkout Loss Prevention” issues an on-screen notification to customers or, if intervention is required, an alert to staff to correct any detected incident before the transaction is completed. This enables recovering or preventing more than 75% of losses attributed to self-checkout, including loss patterns such as skipped scans, items not scanned intentionally, items left in the cart, or abandoned transactions.
“The self-checkout transformed retail, but it also changed the loss equation,” says Christian Floerkemeier, CTO and cofounder of Scandit. “Retailers shouldn’t have to choose between a fast checkout experience and effective loss prevention. The loss prevention system at self-checkout provides a software-based solution to detect issues in real time, with a process configurable to each retailer’s policy. In addition to recovering losses, it also enables retailers to optimize staff utilization more productively.”.
The self-checkout loss prevention system is implemented through a station-based architecture, which eliminates the need for dedicated servers and complex installations, enabling scalable deployment from small retailers to large hypermarkets. This solution, compatible with any hardware, works with existing camera infrastructure (whether integrated cameras, commercial cameras, or standard security cameras located above the self-checkout station). This eliminates the need to use specialized AI hardware and helps maintain a low total cost of ownership.
With privacy as a priority, the solution processes video at the station and is designed to help retailers comply with GDPR obligations. Item and transaction analysis eliminates the need for facial recognition or biometric identification, thus avoiding practices prohibited by the EU AI Act. It is also backed by Scandit’s ISO 27001 certification.
“One of the challenges of visual AI verifications at self-checkout has been the high infrastructure cost and the latency of some traditional solutions, which often created a disconnect between behavior and the moment of intervention, causing confusion among all involved,” adds Greg Buzek, president and chief analyst of IHL Group. “A lightweight solution that’s easy to deploy in stores eliminates that cost and latency, and when existing cameras can be reused, it further reduces costs. This gives retailers an edge: it deters intentional theft and rapidly resolves involuntary mistakes that would otherwise slow the transaction process.”
With this announcement and the July 2026 launch of the Age Verification solution, Scandit significantly reduces the two primary friction points that require employee intervention. “Age-Verified Self-Checkout” enables retailers to automate and verify the purchase of age-restricted items at self-checkout to reduce delays, eliminate unnecessary employee intervention, and improve regulatory compliance. With both solutions in tandem, Scandit offers a more autonomous self-checkout model that helps retailers meet age-verification requirements, protects them from losses, and frees employees to focus on higher-value tasks rather than routine checks.
