Amazon begins rolling out AI-generated review summaries
The paragraph-long blurbs include clickable keywords.
Amazon announced a new generative AI feature today that summarizes product reviews. Available initially to “a subset of mobile shoppers in the U.S. across a broad selection of products,” the artificial intelligence tool creates a recap paragraph highlighting common themes from customer feedback. The company first confirmed in June it was testing an AI-powered summarization tool, but it now begins its official rollout. CEO Andy Jassy said earlier this month that AI is “at the heart of what we do.”
The idea behind the ML-generated summary is to let shoppers get the gist of their peers’ impressions without having to file through a swath of reviews manually. The wrap-up includes a short paragraph describing customer consensus: It’s a bit like an AI-powered version of the “Critics consensus” and “Audience says” blurbs you’d find on Rotten Tomatoes. “Customers like the stability, ease of use, and performance of the digital device,” one example summary shared by Amazon reads. “They mention that it’s way faster, the picture / streaming speed is excellent, and it’s a simple device to get connected. They also appreciate the performance, saying that it performs as expected and works great with LG 3D smart TV.”
The summary is followed by clickable tags showcasing relevant themes and common words from customer reviews. (It’s similar to an existing keyword feature in the company’s reviews.) Clicking on one will bounce you to full reviews addressing the chosen theme.
The elephant in the room is Amazon’s reputation with fake reviews. Although the retailer says it “proactively blocked over 200 million suspected fake reviews” in 2022 alone — and it’s known to sue culprits (and get a hand from the FTC in extreme cases) — that hardly means the company detects and blocks all of them. There’s also the question of whether AI-powered fake reviews (using ChatGPT or similar tools) are more challenging for Amazon to spot than human-written ones.
The company’s strategy includes only unleashing the summarization tool on verified purchases while using AI models that allegedly detect sketchy reviews — and calling in human investigators when needed. “We continue to invest significant resources to proactively stop fake reviews,” Amazon Community Shopping Director Vaughn Schermerhorn said. “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. The new AI-generated review highlights use only our trusted review corpus from verified purchases, ensuring that customers can easily understand the community’s opinions at a glance.”