Starbucks Scraps Failed AI Inventory System After 9 Months
Starbucks has discontinued an artificial intelligence inventory management system after nine months of operation due to the technology's failure to accurately count products and identify stock across its North American locations.
The coffee chain rolled out its Automated Counting software across North American locations in September 2025 through a partnership with technology firm NomadGo. The system was engineered to allow employees to use mobile devices to scan shelves and automatically identify and count products such as milk varieties, syrups, and other supplies—work traditionally completed through manual counting.
When launched, Deb Hall Lefevre, Starbucks Chief Technology Officer, championed the initiative enthusiastically. In a promotional blog post that has since been removed, she stated that “with a quick scan using a handheld tablet, partners can instantly see what’s in stock.” The company positioned the tool as a breakthrough that would free employees from time-consuming inventory work while boosting accuracy and supply chain efficiency.
The deployment, however, produced results far below expectations. The technology frequently mislabeled products and generated inaccurate counts. Store staff reported recurring problems: similar milk types were confused, items were skipped during scanning sessions, and the system’s reliability deteriorated throughout its limited run.
Notably, Starbucks’ own promotional video inadvertently revealed these flaws. The footage showed the technology failing to detect a bottle of peppermint syrup during a shelf scan—a prescient demonstration of the accuracy problems that would ultimately doom the initiative.
Following the decision to discontinue the program, an internal company newsletter informed employees that beverage components and milk inventory would revert to traditional manual counting methods used for other stock categories.
The failed experiment underscores the real-world challenges companies encounter when deploying AI solutions in complex retail environments. While artificial intelligence continues advancing across industries, this case demonstrates the substantial gap that often exists between theoretical capabilities and practical performance in operational settings.
With information from Breitbart News
