Amazon is set to deploy cutting-edge technology to bolster its commitment to quality and sustainability. Dubbed as "Project P.I." (short for "private investigator"), this initiative will be active in Amazon's fulfilment centres across North America, scanning millions of products daily for any defects.

The core of Project P.I. lies in its use of generative AI and computer vision technologies. These advanced tools are designed to spot issues like damaged goods or inaccuracies in color and size before they reach customers. Notably, this AI model not only identifies defects but also digs deep to uncover the underlying causes. This foresight enables Amazon to proactively implement measures to prevent such issues in the future.

The success of this system is evident in its deployment sites, where it accurately pinpoints product flaws amidst the massive volume of items processed each month. Before shipping, every item passes through an imaging tunnel where Project P.I. assesses its condition. If any defect is flagged, the item undergoes further scrutiny to determine if similar products are affected.

Upon identification of defective items, Amazon associates evaluate them to decide whether to offer them at a discount through Amazon’s Second Chance platform, donate them, or repurpose them. This technology acts as a supplementary layer of inspection, augmenting the efforts of manual inspections at various North American fulfilment centres, with plans for expansion in 2024.

Dharmesh Mehta, Amazon’s VP of Worldwide Selling Partner Services, emphasized the company's dedication to customer satisfaction, stating, “We want to get the experience right for customers every time they shop in our store." Leveraging AI and imaging technologies within their operations facilities, Amazon efficiently detects potential product issues, benefiting customers, selling partners, and the environment.

Project P.I. also plays a pivotal role in Amazon’s sustainability drive. By intercepting damaged or defective items before reaching customers, the system reduces returns, minimizes packaging waste, and cuts down on unnecessary carbon emissions from transportation.

Kara Hurst, Amazon’s VP of Worldwide Sustainability, noted the significance of AI in Amazon's sustainability efforts, saying, “AI is helping Amazon ensure that we’re not just delighting customers with high-quality items, but we’re extending that customer obsession to our sustainability work by preventing less-than-perfect items from leaving our facilities." This approach helps mitigate carbon emissions arising from transportation, packaging, and returns processing.

In parallel, Amazon is leveraging a generative AI system with a Multi-Modal LLM (MLLM) to delve into the root causes of negative customer experiences. When defects evade initial checks, this system analyzes customer feedback and fulfilment centre images to pinpoint the source of the issue. For instance, if a customer receives the wrong product size, the system scrutinizes product labels in fulfilment centre images to identify the error.

This technology also benefits Amazon’s selling partners, particularly small and medium-sized businesses that form a significant portion of Amazon’s sales. By making defect data readily accessible, Amazon helps these sellers rectify issues promptly and minimize future errors.

3.5