Phony buyer reviews plague internet as Amazon slams social media giants

Shoppers are inundated with fake reviews and social media sites aren't doing enough to curb the fraud, according to Amazon. 

"Several years ago we saw the rise of what we call fake review brokers," Dharmesh Mehta, head of Amazon’s customer trust team, said, according to the Guardian. Mehta said the phony review scam is a "cottage industry of fraudsters," with hundreds of paid people sending out the reviews to leave shoppers "deceived about what products they should or shouldn’t be buying."

Amazon has faced accusations of letting fraudulent reviews slide, but the company worked to respond last year by taking legal action against 90 brokers who promoted the reviews. It also sued an additional 10,000 Facebook group administrators who tried to post fake reviews on Amazon. The tech and commerce behemoth reportedly uses artificial intelligence to help weed out the fake reviews.

Brokers "go to sellers and on the nefarious side say: ‘I can get you fake reviews’ – but many offer it as a marketing service and end up duping a small business who don’t know what is going on behind the scenes," Mehta said, according to the Guardian. 

"Then they go to a bunch of consumers and say: ‘Hey, if you leave a five-star review for this product, I’ll give it to you for free or a £25 gift card.’ So, they’re effectively buying a customer’s review on one side, and on the other hand, selling a marketing or review service to a brand or manufacturer," he continued.

Amazon reported last year there were more than 23,000 groups promoting fake reviews – which reached more than 46 million members or followers – on social media platforms such as Facebook. Mehta said some social media platforms are not being strict enough when it comes to the phony reviews.

"Some of these groups are advertised as ‘fake eBay reviews’ or ‘fake Amazon reviews,’" he said. "There’s some that will hide, but there’s a lot that are pretty obvious and you’d want folks to be proactively preventing that."

Mehta detailed three steps Amazon is taking to curb the proliferation of the fake reviews in a blog post last week, calling on "all sites that could be used to facilitate this illicit activity" to "have robust notice and takedown processes that are effective and fast."

Amazon has also used AI in recent years to help curb fake reviews on its site and announced the company will invest more funds into the technology to further protect shoppers and sellers. The artificial intelligence systems use algorithms to determine if a review is fake or legitimate, such as the relationship of a reviewer to other online accounts, sign-in activity for a review author, and the author's review history.

"We use machine learning to look for suspicious accounts, to track the relationships between a purchasing account that's leaving a review and someone selling that product," Mehta told the BBC this month.

"Through a combination of both important vetting and really advanced machine learning and artificial intelligence – that's looking at different signals or behaviors – we can stop those fake reviews before a customer ever encounters it," he said.