Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
FTC announces proposed rule banning fake reviews and testimonials (ftc.gov)
217 points by pseudolus on June 30, 2023 | hide | past | favorite | 84 comments


From the proposed rule text [0]:

> It is an unfair or deceptive act or practice and a violation of this Rule for a business to represent, expressly or by implication, that a website, organization, or entity that it controls, owns, or operates provides independent reviews or opinions about a category of businesses, products, or services including the business or one or more of its products or services.

As I read it, the 5-star review systems on Amazon.com and Walmart.com would be categorized as unfair under this rule, since both companies sell their own products in addition to having a rating system on their website. I'm not sure whether that's intentional or not (and I could be misreading it, IANAL).

[0] https://www.ftc.gov/system/files/ftc_gov/pdf/r311003consumer...


No, I don't think that's correct, either by the text or spirit.

This isn't about user reviews, it's about editorial reviews where it is the website/organization itself that is doing the reviews. Think Wirecutter, Consumer Reports, and so forth (although those are fine).

Basically when you search Google for reviews on e.g. an air purifier, you'll find a bunch of sites called things like "air-purifier-reviews.com" that seem to be a legitimate review site/blog at first glance (like Wirecutter), but are just a fake site created by the company whose products are recommended.

Amazon and Walmart with their product listings and user reviews aren't anything like this. User reviews even for Amazon-branded products would continue just fine, although another rule indicates that Amazon couldn't selectively remove negative reviews for their own products (not that there's evidence they have, since there are plenty of terribly-rated ones alongside the good ones).

It's also clear that Amazon wouldn't be able to e.g. create its own Wirecutter or Consumer Reports equivalent as long as it continued its own brands. And it probably means that Amazon wouldn't be able to continue its "Amazon editorial recommendations" which is a row that used to pop up in search results, but that they removed a couple of months ago [1].

[1] https://www.modernretail.co/technology/amazon-has-quietly-re...


That seems to apply only if wire cutter or consumer reports did reviews on wire cutter or consumer reports, isn't that right?


I interpret that more as banning all of the shadier situations like the mattress companies that own “independent” mattress review sites.


I was thinking this but how the same 2-3 vpn providers own all the vpn review websites


I suppose they would be required to work with an independent 3rd party that would manage the reviews. A little bit similar to the Credit Rating companies.

IMHO it could be a good, kind of, separation of responsibility.


> As I read it, the 5-star review systems on Amazon.com and Walmart.com would be categorized as unfair under this rule

Well, in reality they are.


> Businesses would be prohibited from creating or controlling a website that claims to provide independent opinions about a category of products or services that includes its own products or services.

This is literally designed to kill vpn review websites. All the major ones are owned by a couple of vpn mills


I dont think it's designed to kill vpn review websites specifically. it'll sure catch those though.


I’d be surprised if VPN providers were an area of prime focus.

…but wouldn’t this description also include Amazon, with Prime brands at least? It sort of implies a separate site which would mean Amazon is fine, but it doesn’t say it outright.


It will be interesting to see what "controlling" means.

Can they "Sponsor" a site, and put a little tiny disclaimer on the bottom of the page, where no one ever looks, indicating this is sponsored content?


Hell, they could put "Sponsored Content" above the article just like in news readers, and people will think it is legit news


And the sundry video editing tools, pdf editors, random file converters etc. So many just mask their promotion strategies in the form of review blogs.


Mattress review sites, too.


this could work for any number of review sites not just VPN sites


This week I got locked out of my car. I left the keys in the trunk of a 2022 Toyota Camry. A car that happens to have strong anti-theft features.

In a state of desperation, I looked for locksmiths in my area. First Google ad that popped up for “Locksmith Redmond” seemed like a reputable website. To confirm their credibility, I looked up their Google reviews: 400 five star reviews.

I requested the service, and twenty minutes later some dude turns up in a Ford Escape. He proceeds to “open the car” and quotes me $299 on the spot for simply opening the door.

If you’ve dealt with anti theft on these Toyotas, you know this solves nothing. You can’t get into the trunk without disabling it.

I refused immediately: are you insane? He started getting aggressive with me. It’s me and this dude, alone in a parking lot at 10 PM. Dude’s twice my size, I was terrified. Long story short, he didn’t get the money.

It was a scam, a bait and switch. In my state of nerves, I fell for the reviews. I took a look at the reviews after the situation was resolved.

Clearly fake reviews: same writing style, similar names. Only two or three people who actually exposed the scam.

I’m a very tech savvy guy. I got screwed by fake reviews and Google ads this week. I have a feeling this measure won’t stop them.


Locksmiths, plumbing, electricians and other services in any major city are plagued by people whose actual business model is setting up a website, CRM system and pipeline to sell "hot leads" to shady trunk slammer type contractors. It's trivially easy to have a phone number that looks like it belongs to the area code for (your city here) but goes to a call center offshore. It's a whole well documented phenomenon.


Kind of amazed this needs any kind of new rule. Isn't this FTC bread and butter? Still, if it leads anywhere, I'm happy to see the FTC do more work in this direction.


Me too. I'm over here wondering how fake reviews aren't already fraud and why it isn't being actively prosecuted for obvious consumer harm.


I have been asked (even recently) many times by upper management in various private companies, NPOs, and public industries alike to boost local visibility by submitting positive reviews to Yelp, Google, Foursquare, etc. recommending it as a place to work or talking up products/services offered. Often offered a reward for it, in some form or another.

Think of a clerk at your local EBT office submitting a 5 star Google review saying "Excellent customer service and low wait times, [your coworker here] was caring and compassionate, just make sure you schedule an appointment online to avoid the lines!" or "I've worked here for X years and I cannot say enough great things about how wonderful and caring everybody is! I love helping the undeserved in my community!"

Maybe it's a car repair shop doing the same. Maybe it's a fast food franchisee. Maybe it's a boutique store. It's definitely more than just Amazon or TripAdvisor reviews.

I intend to present this proposed rule to my current management and let them sweat, because just this week I was told it is "highly recommended to help offset the fake reviews". I understand the struggle, though. People leave shitty 1 star reviews for spiteful purposes all the time, and that can hurt a small business or encourage unnecessary community discontent.

I don't think there is any good solution on the SEO aspect, but I'm all for banning the scummy practices of astroturfing.


This needed to happen 10 years ago.


I would estimate more like 20.


Honestly, that's a pretty standard "tape delay" for this sort of thing. It's really hard to make good, widely agreed-upon rules before the problem becomes widespread. Once it is, and there's popular support, rules and regulations become obvious.

And that's how government tends to work.


But fraud is already illegal.


Amazon reviews were decent 10 years ago. Now they're worse than useless. My Amazon shopping has followed the curve.

I'm pretty sure Amazon knows best of all what's going on, but stays quiet because they're generating insane profits.


They’ve been broken for a lot longer than a decade.


Oh yes, but now fake reviews are on a an industrial scale.


It's even more devious than that. Some merchants have taken to planting obviously fake reviews of their competitors to get them banned from the platform.


Even going as far as hijacking other's listings!

Crazy story: https://www.kleinbottle.com/Amazon_Brand_Hijacking.html


Now if they'd only ban "this review is based on a free product I got from the company in exchange for reviewing it" fake reviews.


Like others mentioned, disclosed is fine, it's the "you'll receive [$20 | free stuff] if you leave a 5 star review" that makes reviews untrustworthy.


I think those sorts of reviews are fine (both ethically and legally) as long as that is disclosed so that everyone knows to discount the review.


The scores are still being counted in aggregate scores though.


I’m personally conflicted about these. On one hand, having launched products before, it’s difficult to get those first reviews because even if the public perception of reviews is poor, they still hold a lot of weight in terms purchasing decisions. As such, it’s easy to send off a few free products and hope they will review them well. As a consumer, however, I’m always skeptical of products with less than 20 or so reviews that are all 5 stars and the text of the reviews have no real substance.

There is an opportunity here (and I’m not sure what it looks like) to get products in the hands of people who will be incentivized to review honestly because the maker of the product actually wants to earn organic, positive reviews and will take the issues noted by reviewers seriously for improving upon the product.


As long as it is clearly indicated, is it really that big of a problem?

I think they are focusing on getting rid of deception...


Yes, so long as when browsing products there's a checkbox to ignore all such reviews so I can sort by a rating that doesn't include such shenanigans.


Ironically, my landlord has been bugging me for the past few months to review them publicly. This is how they do it: I log in to my tenant portal to check if my account is current, and a full-screen modal dialog blocks my way, promising that I could win a $100 Amazon gift card if only I'd review them on some site I don't know. (And I'd probably need to create an account and jump through all the associated hoops.) So yeah, dudes, I don't use Amazon. Most gift card offers at least provide a choice of two, in case you don't shop with the place. And the dialog is still in my way, months later.

That dialog is the only update they've made to the portal in years. In fact it still doesn't support MFA at all. But reviews!


$100 Amazon gift card + security deposit


This should already be prosecutable as wire fraud. Does an FTC rule make enforcement easier?


i thought the same, then realized phone spammers are out of control still in the US, and they're not slowing down. What would actually work to shut these crooks down?


How about caning. It might seem foreign and extreme, but it works. A fine is just a cost of doing business. Even prison time would usually be spent at a minimum security prison perhaps even with work release. Barely a punishment.

Those are just slaps on the wrist. You know what’s not a slap on the wrist? Caning. Caning would force these scammers to rethink their lives. It is used in Singapore as a punishment, a country widely seen as one of the least corrupt on the planet.


Oh, yes, good idea. Just track down wherever in the world the scammer lives, like Russia or Myanmar or Venezuela or something. Then you fly in a team of Navy Seals in a black helicopter to land at night on the roof of their tenement under cover of darkness. They storm in, restrain the scammer, and cane 'em real good with 39 whacks. Then they beat feet out of there before they can be detected or caught. You could GoPro all of this and then post it to a special "Cane of Shame" website. Exciting!


China, most likely. Good luck sending Navy SEALs without starting WW3.


Could skip all the actual action, and computer generate the same with "AI" and still run the Cane of Shame website.


Caning is barbaric and backwards. It will never happen here in the USA. It would normalize beating people and we already have enough of that illegally via the police


Caning doesn’t normalize beating people any more than fines normalize theft. Far from barbaric, it is a regulated punishment that can effectively deter repeat offenders. And if some portion of prison time were replaced with caning, it would reduce the US’ overflowing prison population. I’m sure you would agree that given how barbaric US prisons are, the less time spent there the better.

And as far as I can tell most of the police beatings are irregular, not regulated, and applied without judicial prescription or medical oversight. American police beatings are also usually legal or at least decriminalized however.


Police in the US at least used to carry batons but I've never seen one used.


This isn't a review of a FTC show. It's not really my genre.

It's a review of that classic 90s sub-genre of FCC, Net Neutrality. This is Net Neutrality 2015: Nettier and Neutrier, with Michael O'Rielly.

Esteemed Chair O'Rielly's performance was fantastic as always. After the entirely authentic acting in the fake-out heist classic between President Obama and Tom Wheeler in the 2014 installment, O'Rielly really steals the show here with his honest portrayal of concern for free speech.

Four stars. A slamming success as always.

I really can't wait for the upcoming FTC show of "Automated Robo-caller Deleters from Outer Space". It will be dealt with in a balanced and classy manner as the posters claim. An honest reflection on space PBX robots and what they experienced.

Do you want a sarcastic parody world? Anyways, this gets us closer to authenticated users everywhere, so that's actually a good thing. Many lives will be saved that day.


I really do understand the plight of victims of scams.

I know of a couple.

I also see people hurt who were chased from communities, or are drawn into these kinds of things right here instead of just working on what they love.


It is sort of amazing how many of the scenarios listed on this page we have all seen, probably on a daily basis.

Wonder how many of these will survive to the end? I guess one good thing is that the participants are not something like ISPs vs the public, more like business vs business.


I worked in a company a few years back where marketing people faked "trust-worthy" software like trust pilot reviews by using everyone phones, or with different IPs. Nowadays a 4.5 average review is not even that good for a product


Trustpilot also has paid subscriptions where the business can pay to send "review invitations". So a shitty business can get a high aggregate score because they've gamed when to ask for reviews.


Most fake reviews come from India and the buyers are using non-US payment systems. I don’t know how this would impact the existing market.


On Amazon at least, they aren't. There are giant networks of companies seeking out American reviewers, basically giving free products in exchange for reviews. I was invited to one after leaving a good review on some product I'd bought, and did some digging.

From what I can tell, Chinese companies hire companies in India and Bangladesh to boost reviews. They then act like middlemen, seeking out pay for play reviewers in other countries. It's funny because they pretend to be Chinese women in their profiles. After chatting one lady 'Eve' up, basically saying I knew they weren't Chinese by the way they wrote English, he admitted to being a young man from Bangladesh, though I forgot his name. I thanked him for his honesty.

I was sent a spreadsheet of multiple items to choose from. Clicked on a few, and saw a couple 'Top 10 Reviewer' labeled folks making reviews on these unheard of products. So the 'all stars' of Amazon are all in on it.

All this is separate from the multitude of companies who put little cards in their packaging offering a rebate for a good review.

Garbage all the way down...


US reviewers often get paid a premium on amazon/iTunes/Apple Store, but the bulk buys are still India.

It’s quite easy to spot fake reviews by anyone, since they’re formulaic. I.e the review must mention <person name> or <product name> exactly as written and be at least 50 characters long. So you get 2-3 sentences that look like this:

“Product is a huge game changer for me and my family. I’ve tried some other stuff, but honestly it didn’t work as well. I’d recommend Product for anyone who has Reason.”


Where does one start working as a paid reviewer anyways?


Sites like mturk. Some of the work seems legit, but it’s all manipulation of organic web traffic. I made a lot of money in fake reviews in a past life.


> On Amazon....There are giant networks of companies seeking out American reviewers...I was invited to one after leaving a good review on some product I'd bought

> From what I can tell, Chinese companies hire companies in India and Bangladesh to boost reviews. They then act like middlemen, seeking out pay for play reviewers in other countries.


This is so non-specific. Which website(s) are you talking about?


Piecework sites. I’ve made thousands of dollars in fake reviews. I made $100 in an hour one time while watching shows on Netflix with my wife. I’ve tried boosting posts on Reddit, but the cost seems high for what you get.

I’m no longer selling anything myself though. I’ve been out of this world for years.


Claiming that millions of people are using a product or service needs to be supported by data. Bandwagon manipulation is worse than individual testimonials.


And meanwhile we let any governing official to tweet whatever they like to their followers.


Rules are fine, but there needs to be a self-sustaining enforcement program.


How will amazon survive this


Always thought there should be a public identity system on the Internet in which your own Internet reputation is at stake like your eBay score. It could be attached to your credit score and if found out your a liar and faker (for any type of personal gain) your Internet identity and credit score takes a hit. Also, You are only able to ever get one just like you are only able to get a social security card and your score/Internet identity follows you around everywhere you go.

Just a thought to clean up all the fake crap that litters the Internet.

Now with this idea you could still post anonymously it just wouldnt hold as much weight.


You could call it a Social credit system.


That's just nopes all the way down for me. It would certainly keep me off of the web entirely (which might not be the worst thing, admittedly).

I'd much rather put up with the fake crap.


How would that keep you off the Internet? You can still post anonymously.


It'd keep me off the web, not off the internet.

If participation is optional, then why would anyone participate? To be effective, it has to be mandatory in some way, be it contractually, legally, or socially.


It might make people be more authentic and real as positive commerce fake reviews is the issue at hand. If im using my U.S. public Internet reputation ID to leave a review then it one says Im in the U.S.(not a foreigner) as ive been granted such an ID in the U.S. and two people would be more prone to be nice. While the negative comments would be in the majority anonymous I'm betting. Though anonymous comments would not hold as much weight but still been seen.

As for an incentive in time especially if a small to mid-size portion of Internet citizens are sick of all the fakeness and start using their Internet public reputation ID then it becomes a societal push.

I think such is needed even more so for all the deepfakes that may proliferate the Internet in the next few years or more. If that becomes a huge problem of really Internet citizens figuring out what is what then maybe that starts the societal push with government offering it but not ever pushing it.


no one is going to sign a legal contract so that they can review crap they bought on Amazon.


Couldn’t Amazon just remove all reviews from their site entirely? I’m sure sales would take a hit, but they could just point to the FTC rule and say “our hands are tied! No more reviews!”

I think a lot of people would continue buying lots of stuff from Amazon. They’d just have to do their product research elsewhere.


?

Amazon would be absolutely thrilled if the FTC successfully cracked down on fake reviews.


Amazon would still be in charge of enforcement, they'd just be held liable for not doing so. Which is not a pleasant place to be, because blocking automation is not an easy problem, and it's difficult to measure success.


"Amazon would still be in charge of enforcement"

Is this true? Based on my reading of the proposed rules, it would be the product owner who would be responsible, not the platform.

I don't think Amazon Basics or any of their other brands are participating in this sort of behaviour.

I could be wrong, though...


> Is this true? Based on my reading of the proposed rules, it would be the product owner who would be responsible, not the platform.

From my reading, Amazon is only responsible for whatever they knowingly participate in. If sellers and manufacturers participate in it without Amazon's knowledge, Amazon has no required enforcement.


> blocking automation is not an easy problem

Amazon is very much willing and able to tackle difficult problems, as repeatedly demonstrated throughout their history. Admittedly they prefer to choose which difficult problems they tackle, but still.


The difficulty of a problem scales in an adversarial environment.


Yes. But the problem is that Amazon themselves would be an adversary to the idea of putting in the required effort for no financial benefit to their business, not that they would be trying to stop other adversaries.

They have solved or attempted to solve plenty of problems just as hard as this one, even acknowledging that this is an adversarial environment. But they generally only choose to do so when the payoff to them is worth the effort. In other words, they can properly tackle it, but they wouldn’t want to.


They will make Amazon do the legwork.


If Amazon has to do the legwork, don't they need a system to identify what's fake to enforce if Amazon isn't doing their part?


Is this going to tackle all paid fake reviews, aka ads?


So... no more free iced tea for a 5 star review?


Is it all affiliates programs actually ?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: