Amazon uses AI for better customer satisfaction and sustainability

Amazon has launched a new initiative called “Project P.I.” (private investigator), which uses computer vision and AI to ensure that customers receive products in perfect condition. This project plays a critical role in Amazon's sustainability efforts, helping to reduce unwanted returns and unnecessary CO2 emissions.

How does Project P.I. work?

At Amazon fulfillment centers in North America, millions of products are scanned for defects every day. Project P.I. uses generative AI and computer vision technologies to detect issues such as damaged products or incorrect colors and sizes before they reach customers. The AI model not only identifies defects, but also helps identify the causes, allowing Amazon to take preventative measures.

Benefits for customers and sales partners

Dharmesh Mehta, Amazon's VP of Worldwide Selling Partner Services, highlights the benefits of this technology: “We want to get the experience right for customers every time they shop at our store. By harnessing AI and product imaging within our operational facilities, we can efficiently detect potentially damaged products and address more of those issues before they ever reach a customer, which is a win for the customer, our sales partners, and the environment.”

Sustainability initiatives

Project P.I. also plays a critical role in Amazon's sustainability initiatives. By preventing damaged or defective items from reaching customers, the system helps reduce unwanted returns, wasted packaging and unnecessary CO2 emissions from additional transport. Kara Hurst, Amazon's VP of Worldwide Sustainability, notes, “AI helps Amazon make sure we're not only making customers happy with high-quality items, but we're extending that customer focus to our sustainability work by preventing less than perfect items from leaving our facilities, and helping us avoid unnecessary CO2 emissions from transportation, packaging, and other steps in the return process.”

Research into negative customer experiences

Parallel to Project P.I., Amazon is using a generative AI system equipped with a Multi-Modal LLM (MLLM) to investigate the causes of negative customer experiences. This system assesses customer feedback and analyses images from fulfillment centers to understand what went wrong. This is particularly beneficial for the small and medium businesses that account for more than 60% of Amazon's sales. By making defect data more accessible, Amazon helps these sellers resolve issues quickly and reduce future errors.

With these innovative technologies, Amazon is taking a big step towards better customer satisfaction and a more sustainable future.

‍