Particle.news
Download on the App Store

Supermarkets Use Algorithms and AI to Decide Who Gets Checked at Self-Checkout

Retail experts say AI risk scores plus camera sensors now steer self-checkout inspections.

Overview

  • A former Albert-Heijn manager and a retail consultant report that most chains use algorithmic systems rather than pure randomness to select shoppers for post-scan controls.
  • Behavioral factors influence selection, including prior missed scans, unusually long shopping times with handheld scanners, basket value and composition, and time-of-day patterns linked to errors.
  • Some retailers have added cameras and motion or weight sensors, including on trolleys or scanners, in countries such as Germany, France, and Switzerland to flag likely misscans in real time.
  • Store employees still initiate and conduct checks, and they may inspect bags when there is concrete suspicion of theft.
  • Separate consumer reporting notes mixed uptake of self-checkout in Germany, with younger shoppers using it more often and the EHI Retail Institute projecting 23,000 to 25,000 terminals in 2025 versus about 5,000 in 2023, alongside frequent user complaints about glitches and false alarms.