Amazon Using Generative AI To Simplify Shopping

Amazon Implementing AI To Tackle Growing Fake Reviews On Products

In recent months, Amazon has been grappling with rampant fake reviews as a result of brokers. In response, Amazon has decided to adopt machine learning models to combat sad reviews.

The company is currently experimenting with a generative AI model that will sum up customer reviews and feedback on a product on the site.

One of the products the generative AI has already worked on is the summary for Apple’s third-generation AirPods. For example, the AI feature, summing up the over 4,000 reviews on the Airpods, wrote that they “have received positive feedback from customers regarding their sound quality and battery life.”

However, the feature also did add info on some of the negative reviews shared by users on the product. “mixed opinions were also expressed about the performance, durability, fit, comfort, and value of the headphones,” it added.

See also

Tech & Science

Reddit Users and Moderators To Boycott For 48 Hours

The AI-generated summary also features a disclaimer warning where the information is from. It reads, “AI-generated from the text of customer reviews.”

However, while the technology does help customers quickly determine the pros and cons of a product, which is also good for the company, the feature does have its limits and demerits.

As with most Large Language Models (LLMs) such as ChatGPT, Amazon’s AI tool is also trained on troves of online data to come up with the expected responses. This brings all the problems that these models have such as concerns about tone, accuracy, bias, and its potential to generate false details.

Reece Hayden, a senior analyst at market research firm ABI Research, questions whether the tool will be able to decipher fraudulent or bot-created reviews. “These reviews will be treated equally and therefore the summary may reflect fake, non-customer reviews,” he said.

This, therefore, brings up the issue of how this method tackles the fake review problem plaguing the platform. If the false reviews are more than actual customer reviews, then this means that they will weigh the AI tool to generate a “fake” summary.

While the company is using AI to tackle the issue, it has also disclosed to the BBC that it is investing in “sophisticated tools” to help flag suspicious behavior in product reviews.

According to Dharmesh Mehta, manager of Amazon’s consumer trust team, “We use machine learning to look for suspicious accounts and to track the relationships between a purchasing account that leaves a review and someone selling that product.”