The Rise Of Fraudulent ‘AI’ Generated Reviews

It was destined to happen sooner or later. In the midst of talk about how “AI” is improving transportation safety for companies like Tesla (or in the case of Boeing airlines, perhaps causing catastrophic crashes), and even helping Facebook censor violent videos from being published on their platform, there is very little talk regarding the seedier side of artificial intelligence (and no, I’m not talking about HD porno).

“People worry that computers will get too smart and take over the world, but the real problem is that they’re too stupid and they’ve already taken over the world.”

― Pedro Domingos

In the world of online marketing, anyway, here’s one thing that’s starting to happen: AI is officially helping scammers get away with more fraud, including fake reviews.

Several days ago, our stalker from California, Naveed Moein (who apparently wants us to make him famous), easily uploaded a fraudulent Yelp review of LittleBizzy using a fake female profile photo generated using AI. Ostensibly, the reason for this was that nobody could back-trace the photo to a real person.

Using his background in IT, Moein was aware that Yelp’s algorithms would likely flag his fake review of LittleBizzy unless he primed the account. So, first he wrote 2 generally positive reviews of restaurants in California’s Bay Area. On his third review, he (falsely) accused me of sexually harassing this non-existent “woman”, under a fake name, and a fake AI-generated profile photo, and Yelp published it without a second thought.

The photo was generated from a recently trending website, ThisPersonDoesNotExist.com, which is an experimental tool created by Nvidia a few months ago to show off their latest computer chip wizardry.

Despite all 3 of his “reviews” on Yelp being posted within a few hours of each other, and despite LittleBizzy’s Yelp profile being in Las Vegas, Nevada, and the other 2 restaurants Moein reviewed being from the San Francisco, California, area, all reviews remained online — at least, for a while.

Eventually, after a few days, Yelp’s algorithms did in fact downgrade Moein’s profile, and after we alerted their team to what Moein was up to, they finally went ahead and shadowbanned the account.

But still, it raises a pretty interesting question. If Yelp, whos anti-fake reviews software is arguably the best in the world, can’t even stop artificially generated face photos, what happens when AI bots get good enough that they can also register new accounts, choose a name (hopefully more clever than Moein’s choice of “Samantha ‘Sammy’ L” for a Millennial white chick), and all profile information by themself too?

Using malware as an entry point to “normal” looking computer or website IPs around the world, this could rapidly launch a new era of fake consumer reviews…

Note: Moein, who was allegedly fired from his job after his company found out about his non-stop harassment of our team, is understandably upset. After all, in less than 2 weeks, we were able to rank #1 on Google for searches of his name, warning other online agencies of this scammer’s tactics. Despite giving him several chances to stop his behavior, Moein has chosen to press on, slandering our company with fake stories on the likes of Reddit, Facebook, Bark, Serchen, and other online forums and consumer review websites, despite never having been our customer. And yes, while we could easily sue the young man, it’s just not worth our time (at least, for now), especially when SEO and the public square are much more effective — and transparent.