Amazon announced this morning it will begin leveraging generative AI to help customers better understand product feedback without needing to read through dozens of individual reviews. The retailer stated it will use this technology to generate a short summary paragraph directly on product detail pages, highlighting key features and customer sentiment derived from existing reviews.
This text snippet is designed to help shoppers grasp common themes across reviews more efficiently
In addition to the summary, Amazon will highlight specific product attributes as clickable buttons. For instance, if a customer wants to learn about a product’s “ease of use” or “performance,” they can tap a button to filter and view only the reviews that mention those specific topics.
Amazon previously offered a similar feature that surfaced frequently used words from reviews, which were also accessible
The new AI-powered tools will initially roll out to a subset of U.S. mobile shoppers across a “broad selection” of products, the company said. During this testing phase, Amazon plans to refine its AI models to improve their accuracy. The company is also working to expand these highlights to additional categories as the feature becomes more widely available.
Of course, these AI summaries are only as reliable as the data they process. Amazon has struggled for years with misleading and fake content, including paid reviews.
In 2021, for example, the company reported that it had blocked 200 million fake reviews the previous year. It has also taken aggressive action against bad actors, including suing sellers who purchased fraudulent feedback. Last year, the company sued the administrators of 10,000 Facebook groups involved in brokering fake reviews.
More recently, the FTC intervened, forcing a supplement manufacturer to pay $600,000 in a case involving “hijacked” Amazon reviews—a practice where different products are merged into a single listing to boost one item’s rating using the positive reviews of another.
As AI capabilities grow, fake reviews may become increasingly difficult to identify, as the technology can now mimic human writing styles. This could lead to a surge in fraudulent content, potentially undermining the utility of Amazon’s AI summaries unless the company implements robust methods to keep AI-generated spam off its platform.
Amazon addressed these concerns today, stating that its summaries will only incorporate reviews from verified purchases. Furthermore, the company continues to invest “significant re
“This includes machine learning models that analyze thousands of data points to detect risk, including relations to other accounts, sign-in activity, review history, and other indications of unusual behavior, as well as expert investigators that use sophisticated fraud-detection tools to analyze and prevent fake reviews from ever appearing in our store,” notes the retailer.
One thought on “Amazon taps generative AI to enhance product reviews”