The company announced that it is using an AI model called ‘Project PI’ (private investigator), which uses generative AI and computer vision technology to flag problems in its fulfillment centers with customer orders.
Specifically, ‘Project PI’ is able to find faulty or damaged products, or flag items that may be the wrong size or color, while products are processed in the company’s imaging tunnels, where they are scanned and evaluated. Damaged items are reviewed by Amazon employees, who then determine whether they can be sold at a discount on Amazon’s Second Chance platform.
The goal is to improve the overall quality and accuracy of Amazon’s shipping, which could reduce returns and help the company create a more sustainable shipping system.
Amazon also says it is using generative AI systems to figure out what went wrong when items are damaged or incorrectly included in a customer’s order. This, once again, can help the company solve problems in its massive customer service system, making it more efficient and sustainable.
Project PI isn’t the only AI tool Amazon is leveraging to become more environmentally friendly. In recent years, it has also incorporated models that help reduce packaging use, monitor product quality through its grocery service, and use algorithms to estimate a product’s carbon footprint.
Additionally, Amazon Web Services (AWS), the company’s cloud services arm, is also betting heavily on AI. Last year, it launched a service that offers access to AI models from several companies.
*With information from Fast Company/Cover photo: iStock
Follow Adnews on Instagram e LinkedIn. #WhereTransformationHappens
