Robust Proactive Controls
We use advanced machine learning and graph neural networks to proactively monitor all aspects of our store for potentially fraudulent, infringing, inauthentic, non-compliant or unsafe products or content. These systems operate continuously throughout every step of the process – from the moment someone tries to register a new selling account, create a listing and update a product listing. Amazon continually incorporates the feedback we get from customers, brands and others.
Amazon uses advanced technology and expert human reviewers to verify the identities of potential sellers. Prospective sellers are required to provide a variety of information, such as government-issued photo IDs, taxpayer details and banking information. We employ advanced identity detection methods such as document forgery detection, advanced image and video verification, and other technologies to quickly confirm the authenticity of government-issued identity documents and whether they match the individual applying to sell in our store. In addition to verifying these, Amazon’s systems analyse numerous data points, including behaviour signals and connections to previously detected bad actors, to detect and prevent risks.
Preventing unsafe products from getting into the hands of customers begins the moment a new seller starts registering an account with our robust, proactive seller vetting controls. Once sellers are verified and products are listed, our advanced machine learning technology scans product listings and customer interactions weekly, for signals of abuse or safety concerns. We investigate, identify and remove products from our store if a concern does arise or we learn of a recall. We work with sellers, manufacturers, brands and government agencies to act quickly and, if necessary, put new proactive controls in place.
From the moment a seller attempts to register a new account or a product is listed for sale in our store, our advanced machine learning technology is scanning for potential counterfeits, fraud and other forms of abuse. These machine learning models continuously ingest and learn from new information stemming from valid notices of infringement or customer feedback, which helps make our technology even more effective at identifying bad actors and bringing counterfeits to zero.
Amazon’s automated technology scans billions of attempted changes to product detail pages daily for signs of potential abuse, including the creation of new listings and changes to existing listings. Our tools use advanced machine learning to prevent the attempted listing of counterfeit or infringing products – scanning keywords, text and logos which are identical or similar to registered trademarks or copyrighted work.
Amazon welcomes authentic reviews – whether positive or negative – but strictly prohibits fake reviews that intentionally mislead customers by providing information that is not impartial, authentic or intended for that product or service. Our machine learning models analyse thousands of data points to detect risk while our expert investigators use sophisticated fraud detection tools to analyse and prevent fake reviews from ever appearing in our store. When we strongly suspect that a review is inauthentic, we suppress the review completely, so it is not displayed.
