Introduction With online retail sales exploding in recent years, especially during the pandemic, growing to 14 percent of consumer purchases, shoppers are increasingly relying on online customer reviews. Such reviews have skyrocketed, but are becoming less and less trustworthy. Up to 45 percent of online comments are estimated to be posted by people who have not tried a product or service or are paid—in cash or in kind—to give a thumbs up or, to hurt a competitor, a thumbs down. Some firms suppress bad reviews. Such bogus or hidden endorsements cost consumers an estimated $152 billion annually. For business, they pose unfair competition and, ultimately, engender a loss of trust that drags down online sales. Consumers should beware of overly rosy reviews, particularly those using vague terms and posted in groups around the same date. Fraudsters, however, are ever more sophisticated, as is computer-generated text, prompting calls for better oversight by federal and state officials, who struggle to keep up. Online customer reviews have skyrocketed, but the proliferation of fake comments from people paid to boost or denigrate a product or service makes such reviews untrustworthy. Fake reviews cost consumers some $152 billion annually and pose unfair competition for businesses. (AP Photo/Jenny Kane) | Go to top Overview Several dozen people recently gave 5-star reviews on Yelp to a small California home-remodeler, praising the company, for example, for capturing a “modern Ikea vibe look.” But at least one reader does not buy it — not just the worthiness of the company but also the authenticity of the reviews. A Tripadvisor sign on the door of a Nashville, Tenn., business invites customers to review the establishment. Some 95 percent of travelers say they check online reviews before booking. Favorable reviews can provide a significant economic boost to a business. (Getty Images/Robert Alexander) | “Is it reasonable to believe that 35 college-age kids from Southern California all use the same Bay Area home contractor some 400 miles away?” asks Kay Dean, a former criminal investigator who, after being duped herself by a misleading review of a doctor, spends her time exposing fake online reviews. Technology experts — academic researchers, consultants, consumer advocates and federal and state regulators — caution that online reviews are increasingly fake. They estimate as many as 45 percent of all online reviews are bogus, providing false narratives on everything from restaurants, doctors, headphones, clothing and home repair to body creams and vacuum cleaners. Hiding negative reviews is another misleading practice. In other cases, reviews might be written for one product and “hijacked” to use for another. Anyone reading Amazon reviews of consumer electronics only to find them praising photo frames and throw blankets will recognize the phenomenon: A seller accumulates positive reviews for one product, then, instead of creating a new listing, deliberately edits the previous listing to offer an unrelated product while retaining the now-misleading positive reviews. Consumers are becoming more aware of fake reviews online. The number of Trustpilot reviews flagged by consumers in 2021 rose to 110,663 from 88,412 in 2020. In 2021 the number of organic reviews rose 17 percent. Most consumers believe 5-star reviews and 1-star reviews are most likely to be fake compared to 2-, 3-, and 4-star reviews. Source: Laurie Sullivan, “Fake Reviews Surged In 2021,” MediaPost, May 20, 2022, https://www.mediapost.com/publications/article/374082/fake-reviews-surged-in-2021.html Data for the graphic are as follows: Review Rating | Percentage Removed | 5 Stars | 46% | 4 Stars | 8% | 3 Stars | 8% | 2 Stars | 8% | 1 Star | 29% | Fake and misleading reviews cost U.S. consumers an estimated $152 billion a year, researchers say. That cost includes items not delivered or of low quality or that otherwise differ from their advertisement. It also includes lost time and money from what economists call the “opportunity cost” of having less information about better or more appropriate alternatives. Inauthentic reviews also victimize hundreds of businesses who follow the rules only to lose customers to those who game the system to push themselves up — and competitors down — in search-engine rankings. And fake reviews can be dangerous by steering people to less competent or even unqualified doctors, heating repairers or others providing services that impact health. Rules and Regulations Publishing a fake review is illegal under Federal Trade Commission (FTC) rules prohibiting false advertising. The commission in recent years has stepped up efforts to police false and deceptive advertising and marketing, including fake and misleading reviews, often among the biggest players. However, hucksters outpace the regulators, aided by a recent U.S. Supreme Court ruling that curbed — but did not eliminate — the FTC's ability to impose financial sanctions on wrongdoers. The agency can still win injunctions that order bad practices to stop, but it now has a harder time imposing fines in these civil suits without first stepping through some hoops, including having first issued a warning to the company before the suit. In response, the agency has issued warnings to hundreds of companies, even those it has no reason to believe are manipulating reviews. It is pondering new rules and increasingly taking legal actions against fake reviewers, in tandem with state regulators who have more leeway to impose monetary penalties. “The good guys in business are getting hurt by the bad ones,” says Teresa Murray at Public Interest Research Group, a nonprofit consumer advocacy organization. “Fake reviews undermine trust in everyone.” Only half of shoppers now trust online reviews as much as they do word-of-mouth recommendations, down from 84 percent in 2015, one survey found. Consumers are more wary of positive reviews than negative ones. Ongoing media coverage of platforms such as Tripadvisor and Amazon having issues with paid positive reviews has resulted in skepticism from consumers. Source: Jamie Pitman, “Local Consumer Review Survey 2022,” Brightlocal, Jan. 26, 2022, https://www.brightlocal.com/research/local-consumer-review-survey-2022/ Data for the graphic are as follows: Reason for Mistrust | Percentage | Over-the-top praise | 45% | One of many similar reviews | 40% | Anonymous or pseudonymous | 38% | Over-the-top criticism | 36% | One of the only positive reviews | 32% | Just stars, few words | 31% | Seller says it's fake | 26% | Uses a lot of capital letters | 25% | Companies may legally remove posts that are threatening or contain inaccurate or private or proprietary information. But they cannot suppress honestly held opinions or promote fake ones, a category that includes commentary that has an undisclosed bias, is bogus or has been otherwise manipulated to mislead. Perpetrators of these deceptions typically: Have never used a product or service, or Fail to disclose they have been paid in cash or given a free product in exchange for posting comments, or Agree — through Facebook, other social media or less formally — to swap good reviews or bad reviews about competitors with others in an I'll-scratch-your-back-if-you-scratch-mine arrangement. These categories often overlap, creating myriad conflicts of interest that taint information and distort free markets. Regulators, researchers and consumer advocacy groups say consumers should not only subscribe to the creed of “buyer beware” but should be wary of reviews they use to weigh purchases. Doctoral student Rajvardhan Oak, part of a team at the University of California, Davis, researching fake reviews, infiltrated the complex web of this industry, which hides in plain sight. Responding to a solicitation, he agreed to write a good review of a neck pillow in exchange for being reimbursed for buying one. He then was sucked into a world of solicitations to skew reviews. His research team identified three key groups: sellers based domestically or, often, when selling cheaper products, in China; agents, often operating from overseas in countries such as Pakistan and Bangladesh, who recruit reviewers; and “jennies,” as agents call reviewers who, for a price, post inauthentic comments. Professional Fakers Tech writer Rob Pegoraro says the agents in this ecosystem, known as “reputation defenders,” span the spectrum from legitimate to not. And the jennies, Oak says, range from unscrupulous retailers who swap fake reviews among themselves to people simply looking for extra income. Posting fake reviews or suppressing bad ones raises a firm's ranking — and nets it more eyeballs — enabling it to gain traffic to sell more goods and services via key online platforms, from Yelp, Google and Trustpilot to Amazon, Walmart and Target. Research suggests even a 0.4 increase in a 5-star rating system, from 4 stars to 4.4 stars, for example, significantly raises a product's ranking from the bottom half of items to the top 6 percent, making it likely that many more viewers will see it without having to scroll down. And that results in more sales, which researchers and regulators say provide a powerful incentive to violate state and federal rules barring deceptive advertising. The cost of one positive fake review can yield purchases worth many times that. An extra star on Yelp, for example, can increase sales from “5 to 9 percent.” In one case, federal regulators sued a firm that made $5 million selling DVDs of learn-on-your-own guitar lessons, revenue amounting to five times what the company paid for positive endorsements. “Reviews can make or break a purchase,” said a blogger for Insider Intelligence, an e-commerce consulting firm. “They can get shoppers to spend more than they initially intended to” or, with a bad review, steer sales away. Lina Khan, chairwoman of the Federal Trade Commission, testifies before a House subcommittee in May 2022. She said online reviews are essential for businesses, but it can be difficult to know when they can be trusted. (Getty Images/CQ Roll Call, Inc./Tom WIlliams) | Online commerce, among the fastest growing sectors in retail, represented 14 percent of the estimated $6.5 trillion in U.S. retail sales in 2021 and is predicted to reach more than 18 percent in 2023, triple the percentage a decade ago. A simultaneous rise in fake online reviews has left many consumers angry, not to mention inconvenienced and often poorer from products and services that do not measure up — from outdated electronic toothbrushes and advertised rental apartments that don't exist to unrecognizable clothing, poor doctor care, faulty appliance repair and subpar restaurant food. Nearly 8 out of 10 online shoppers read reviews, with the rate approaching 100 percent for local businesses. Unsurprisingly, only a low single-digit percentage of customers patronize businesses with 2 or fewer stars out of 5. One study finds that even when reviewers disclose ties that pose a potential conflict of interest, consumers often disregard how such ties influence positive ratings. “Reviews are essential, but it's hard to know when they can be trusted,” said FTC chair Lina Khan. Fake ones “pollute the marketplace and put honest businesses at a competitive disadvantage.” Iain Taylor, who told the BBC that he wrote fake reviews for companies to earn extra cash, says he would never rely on an online review for anything. (BBC News/Screenshot) | The remodeling firm that the California students endorsed elicited this comment from a real customer, who says the company did poor quality work: “Stay away from these crooks.” Iain Taylor, a middle-aged man in England who wrote fake reviews for extra money, told BBC News: “I have written reviews from numbing creams to eBooks to downloadable independent films. I've learned never to rely on reviews for anything.” Independent reviewers such as Consumer Reports provide meaningful pros and cons of big-ticket items such as automobiles, washers and dryers and computers. Likewise, groups such as Washington, D.C.'s Consumer Checkbook provide reliable reviews of local plumbers and other regional service providers. But billions of dollars worth of smaller-ticket items and local services lack independent assessments, rendering online reviews for hundreds of retailers ripe for abuse. Ever smarter computers with software that uses artificial intelligence (AI) compound the issue by firing off human-sounding missives that fake-review police find hard to spot. The result is an electronic game of fake review Whack-a-Mole. Once identified online or in lawsuits, dishonest players typically disappear quickly from Facebook and Yelp, only to reappear under a new name, and, often, from an offshore location. Platforms such as Google, Yelp and Amazon have programs designed to ferret out the fake from the real, but the mounting number of bad actors often defeats those efforts. Even platforms, such as hotel and travel bookers who earn fees by referring customers, have incentives to maintain a good reputation in order to build trust and generate repeat business. And many companies rely on honest reviews to improve their goods and services. Renate Holland, the owner of several fitness centers in the Munich, Germany, area, sued Yelp, the online review app, for its rating system, saying she lost customers because of a poor overall rating. (AP Photo/picture-alliance/dpa/Lino Mirgeler) | Nevertheless, some of these companies stray into phony review territory when it comes to selling their own products. The FTC and several states, for example, just sued and settled with Google for soliciting endorsements for its Pixel 4 cellphone from radio broadcasters who had never used it. Despite such efforts, “there's a lot of incentive to cheat and not as much to catch the cheaters,” says Daniel Castro, director of the Center for Data Innovation, a nonprofit focusing on technology and internet policy. Go to top Background The History of Hype Misleading advertising has been around forever. Erik the Red, the legendary Viking, promoted settling in an ice-encased landscape in the late 900s by naming it Greenland. Authors occasionally buy scores of their own books to raise their ranking on bestseller lists. Bribes known as “payola” or “pay-to-play” arrangements have been rampant for decades in the music business to get artists' songs prominently aired on the radio. To influence what shoppers notice and buy, food and beverage sellers routinely pay grocers for prominent placement on shelves and, along with myriad other product makers, vie to appear in TV shows and movies. Erik the Red, a 10th century Norwegian navigator, was an early adopter of false advertising: he promoted an ice-encased place by calling it Greenland. (Getty Images/Bettmann/Contributor) | At the end of the 1800s and early 1900s, as public concern grew about the potential public harm caused by proliferating monopolies in energy, steel, railroad, sugar and other industries, Congress enacted antitrust legislation to break up companies and police anticompetitive practices. As part of that mission, which emerged as America was transforming from an agrarian to consumer-driven economy, Congress created the FTC in 1914 to protect against “unfair methods of competition” and “unfair or deceptive acts or practices. That mission included ensuring that ads are “truthful,” meaning they “cannot be deceptive or unfair, and must be evidence-based.” The FTC's authority includes ensuring that financial transactions comply with truth-in-lending laws. And while courts have held that the First Amendment of the U.S. Constitution generally protects what companies can say, fraud or deceptive advertising is not protected. Dawn of the Internet The rise of online reviews coincides with consumers' use of the internet to shop. Growth began slowly in the mid-1990s with the advent of platforms, such as Amazon in 1994 and eBay in 1995, and climbed as more and more people gained access to the Web. Household use of the internet more than doubled in the United States from 18 percent in 1997 to 42 percent by 2000 and then broke 50 percent in 2001. After the dot.com stock-market bubble, which started in 1998, burst at the end of 2000, growth slowed but nonetheless steadily continued: 61 percent of households had internet service by 2007, and by 2019, nearly 87 percent did. Members of the Federal Trade Commission, formed in 1914 to police deceptive and fraudulent corporate advertising practices and encourage competition, pose for a photo. (Getty Images/Universal Images Group/HUM Images/Contributor) | Until the late 2000s, reviews for restaurants, movies, appliances and cars had been the purview of professional guides such as Zagat or newspaper and magazine reviewers or word of mouth from neighbors or co-workers. Then, as reviews became more common on Amazon and eBay, smaller review sites were founded in the early 2000s. Many were quickly purchased by bigger internet players such as Google. Others founded in this period became household names by 2015 — as internet use soared — and endure. Tripadvisor, for example, an online travel agent with customer reviews as a central feature, was founded in 2000. Yelp was created in 2004 to help people find all kinds of referrals and evolved to a site with restaurant reviews as a main feature. As the internet has grown over the last 40 years, so has online retail shopping, especially in the last 15 years, a trend augmented by COVID-19. Marketers long have known — decades before the internet — that word of mouth provides the most effective advertising. With the advent of social media and its influencers (famous social media personalities who can affect who buys what, when, especially among the 20-somethings market), online word-of-mouth campaigns — endorsements and reviews — have become an essential part of any company's web presence. This has led over the last 10 years, especially the last five, to an exponential growth in retailers' online surveys and requests for buyers to rate their purchases and share those reviews publicly. This pervasiveness has led to “survey fatigue,” with fewer and fewer authentic reviews being freely given. More than a decade ago, HBO comedian Bill Maher joked on his show about one company: “I was actually pretty happy with your customer service, up to the point where you asked me to take a survey about your customer service.” Customers' increased reliance on online reviews — even as they increasingly resist providing them and disbelieve what they say — has fueled sellers' incentives to engage in underhanded manipulation of testimonials. One tactic that has become more common is for firms to threaten customers who publicly complain about poor service or products in the hope that even if what the reviewer says is true, the cost of litigation will deter them from speaking out. Such lawsuits are known as SLAPP suits — an acronym for a Strategic Lawsuit Against Public Participation. Legal Landscape Several states have passed protection laws in response, though when it comes to the sale of products and services, state attorneys general and FTC officials say existing law already makes it illegal to try to suppress honest reviews. (However, the law has not always been on the reviewer's side. In the 1990s, “food libel” laws enacted by more than a dozen states were designed to prevent consumers from making disparaging comments about the safety of their food products. Oprah Winfrey was famously sued by the Texas Cattlemen's Association under such a law, after she did a show on the danger of mad cow disease. The cattlemen lost.) In 2016, President Barack Obama signed into law the Consumer Review Fairness Act, which prohibits firms from requiring customers to sign contracts promising not to write negative reviews. Both the FTC and states enforce the law, and states can bring additional charges under their own laws prohibiting unfair and deceptive marketing. Thirty-two states and the District of Columbia, for example, have statutes to protect against lawsuits intended to intimidate and silence individuals from speaking freely, including by publicly complaining about a product or service. As the FTC stepped up its efforts to fight bad practices, the U.S. Supreme Court in 2021 issued a ruling that makes it harder for the agency, when seeking an injunction to stop an alleged fraud, to also collect money from a firm to repay consumers who the FTC says are hurt by the company's fraud. The case was a victory for payday lender AMG Capital Management, which had sued the FTC on claims the agency exceeded its authority in trying to make the company repay more than $1 billion of ill-gotten gain. The ruling preserves the FTC's ability to bar companies from charging customers using deceitful tactics but makes it much harder for the agency to continue to rely on, as it increasingly had over that last decade, “restitution and disgorgement” to discourage fraud by making it less profitable. The agency, for example, won “almost $12 billion in 2016 alone, including $10 billion in a settlement with Volkswagen” over misleading statements about emissions. The unanimous ruling by the Supreme Court reflects an acknowledgment by all the justices, even those who one might expect would favor giving the FTC more power, that current law is ambiguous and only new legislation from Congress can clarify that. It also marked the latest victory in a decades-long effort by some companies and individuals to push deregulation by using the courts to curb regulators' powers, not just at the FTC, but at agencies overseeing banking, securities markets, the environment, housing, education and health care. The U.S. Chamber of Commerce, the largest and most powerful business lobby, has spearheaded industry's decades-long efforts to curb the power of the FTC. Most recently it has launched a campaign to fight what it calls the FTC's “war” on “free enterprise,” specifically the agency's efforts under President Biden to more actively police antitrust violations and deceptive marketing. The U.S. Chamber of Commerce, the nation's largest business lobby, has led efforts to curb the FTC's power and its alleged “war” on “free enterprise.” (Courtesy Almonroth) | Besides public pronouncements, in July the Chamber filed a suit against the agency that is still ongoing to try to force it to produce documents the group has requested under the Freedom of Information Act on how commissioners have reached decisions. And it has helped stall any potential new law by urging lawmakers on both sides of the aisle to oppose legislation that would bolster the agency's powers by, for example, reversing the Supreme Court ruling that restricted its ability to impose monetary remedies. The ruling sparked calls from the FTC and consumer advocates for Congress to explicitly allow the FTC to more easily require companies to repay ill-gotten money to consumers. After the court's ruling, legislation was introduced by Rep. Tony Cárdenas, D-Calif., in 2021, which passed the House of Representatives and died in the Senate. It has not been reintroduced since. Go to top Current Situation FTC Workarounds Despite the lack of congressional action, in response to the Supreme Court's ruling, the FTC has taken several steps to enable it to continue to impose fines. Relying on different rules than those used in the Supreme Court case, the FTC began using its authority to collect fines if it gives companies advance notice. It sent letters to hundreds of companies warning that they could face “steep penalties” for using “fake reviews” or “other misleading endorsements” to “deceive consumers.” This “notice of penalty offenses” cited administrative law cases to tell firms they “could incur significant civil penalties” of nearly $44,000 per violation. That could run into the millions of dollars depending on how often and how long a company touts a fake review, hides a bad one or engages in other deceptive behavior. A year later, on Oct. 20, 2022, the FTC announced that it will consider proposing new rules “to combat fake reviews and other deceptive endorsements.” It will review public comments received until early January before deciding. In January 2022, the FTC settled its first case on charges that a firm suppressed bad reviews. The firm, online clothing retailer Fashion Nova, agreed to pay $4.2 million based on claims it had blocked customers' criticisms. The agency also has beefed-up efforts to join legal actions brought by states, which can more easily impose fines. In August, for example, the FTC joined California, Colorado, Florida, Illinois, Massachusetts and New York in suing rental company Roomster on allegations that it flooded its online listings with phony reviews the FTC calls “testi-phony-als.” States and the FTC also linked arms to crack down on fraudulent ads on the radio. In November, the agency joined seven states in settling with Google and iHeartMedia, the largest radio network in the United States, over allegations that Google paid to have its Pixel 4 cellphone endorsed by broadcasters who never used it. The FTC and the states say radio hosts 29,000 times read scripts such as “It's my favorite phone camera out there, especially in low light, thanks to Night Sight Mode,” “I've been taking studio-like photos of everything,” and “It's also great at helping me get stuff done, thanks to the new voice activated Google Assistant that can handle multiple tasks at once.” However, the FTC said, “the on-air personalities were not provided with Pixel 4s before recording and airing the majority of the ads and therefore did not own or regularly use the phones.” The companies agreed to stop the misrepresentations and to pay Arizona, California, Georgia, Illinois, Massachusetts, New York and Texas $9.4 million in penalties and costs. States also have become more active on their own. In January, for example, Washington state sued a regional plastic surgery practice on allegations that it violated state and federal law by posting fake reviews, requiring patients to sign agreements promising not to post negative reviews and threatening and bribing patients who publicly criticized the practice. “Threatening and bribing customers to prevent them from sharing the truth about their experience isn't just wrong — it's illegal,” Washington Attorney General Bob Ferguson said. Private Firms Take Action Online retailers and providers of reviews also feel pressed to do more about fake reviews. “The prolific fraud and deception … is severe enough that on a near-annual basis, e-commerce giant Amazon announces yet another crackdown,” the law firm of Brownstein Hyatt Farber Schreck told clients. Amazon did just that last July, filing a high-profile suit against those who it said run more than 10,000 groups on Facebook organized to post phony reviews on the online seller's site. It also has booted off its site several firms for violating rules requiring that reviews be authentic. In mid-2021, for example, it jettisoned the brand Aukey, a popular maker of charging devices and other electronics. The review website Trustpilot, in a bid to enhance public trust, has taken steps to be more open about its business by issuing a “transparency report.” Last year, in its second such report, it said reviews on its site had risen 21 percent in 2021, to 46.6 million, of which it removed 2.7 million, or 5.8 percent, for being fake. Similarly, Yelp reports that in 2021 it removed 8 percent of its 244 million reviews. One remedy would be for platforms posting reviews to require that reviewers identify themselves, at least to the company, even if their name and contact information are hidden from the public. Trustpilot said 85 percent of people it surveyed in the United Kingdom and the United States say, “they would be willing to prove their identity to protect others from misinformation online.” But, says researcher Oak, “There's enough legitimate activity that it can hide the illegitimate activity,” making it “hard for Amazon” or anyone to detect. Advances in artificial intelligence complicate the issue by making computer generated material harder and harder to distinguish from human writing. (See Short Feature.) Go to top Outlook Reforms Needed Technology experts agree that federal and state regulators should intensify efforts to combine resources in court and work with review and retail sites to establish best practices and share information. They also say Congress should strengthen legislation to protect posters of “honest reviews” from retaliation, going beyond the 2016 federal law that bars firms from forcing customers to agree in writing not to criticize the companies publicly. That would ensure customers in all states are protected, not just those in the 32 states with such laws. The Wall Street Journal recently highlighted the story of an Amazon customer who panned a cooking-oil sprayer that she said: “was like a Super Soaker gun … which defeats the purpose of the product.” The seller, in violation of Amazon's rules, contacted her directly, asking her to delete her comments. Amazon eventually removed the seller from the site, but only after the customer's repeated requests. Many consumer activists also want Congress to revamp the law that shields platforms from liability for content published on their sites, which proponents say would propel sites to more actively police their reviews. The White House and some Republicans favor adjusting the shield, known as Section 230 of the Communications Decency Act of 1996, but they do not agree on why or how to do so. The U.S. Supreme Court might act where Congress has not, however. In February the court heard arguments on whether platforms such as Google should be held liable for content promoting terrorism. A ruling, expected later in the year, could reshape the internet by holding internet conduits such as Google responsible for the messages and information sent over their platforms and thus require them to police for dangerous or false communications. With or without a landmark decision by the Supreme Court, legislative remedies are not likely soon. The House is controlled by Republicans, who in turn are controlled by a small but vocal group eager to curb, not expand, federal power, including at agencies such as the FTC. These forces also could imperil the agency's funding, stifling its ability to maintain current enforcement levels, let alone expand efforts. Industry often lobbies to cut the budgets of federal regulators, hampering oversight of businesses ranging from Wall Street to Silicon Valley to fossil fuel producers. Industry, emboldened by the Supreme Court ruling curbing the FTC's ability to impose financial penalties, also continues to push lawsuits designed to curtail the power of regulators. Two cases the Supreme Court likely will decide in June could further curtail federal agencies. One involves the Environmental Protection Agency, the other the Securities and Exchange Commission. Although the two cases are not related to the FTC specifically, decisions on the reach of any administrative agency can affect how courts will rule on future challenges to federal oversight. Others would like for platforms to be more transparent, such as by announcing when they remove reviews. Lawmakers could include such a policy in legislation, but platforms could also do this voluntarily. A non-legislative remedy, says activist Dean, would be for the FTC “to begin a public relations campaign to warn consumers about the prevalence of fake online reviews and encourage them to get business referrals from real people.” Go to top Pro/Con Pro Chairperson, Federal Trade Commission. Excerpted from the FTC's November 2022 Advance Notice of Proposed Rulemaking and Request for Public Comment | Online shopping runs on reviews. When you're in a brick-and-mortar store, you can see the inventory. If it's a couch, you can sit on it. If it's a TV, you can watch it. But when you're shopping online, it's much harder to know what you're actually buying. That's why reviews are so crucial. If 500 other people have bought something and say it works, you can have a lot more confidence. But what if those people were paid to leave those positive reviews? Or what if they're bots? What if the seller is hiding a thousand one-star reviews? That's the dilemma when you shop online. Reviews are essential, but it's hard to know when they can be trusted. Precisely because of the importance of reviews, firms can face powerful incentives to game the system. Businesses have been caught leaving positive reviews for their own products or services, suppressing negative ones, and boosting bad reviews of their competitors. The incentives extend beyond the seller of the product itself. The platforms that host reviews may also, in some instances, benefit indirectly from fake ratings and endorsements and have financial incentives to turn a blind eye to misconduct that brings in revenue. These practices don't only harm the consumers who place their trust in fake reviews. They also pollute the marketplace and put honest businesses at a competitive disadvantage. | Con Commissioner, Federal Trade Commission. Excerpted from the FTC's November 2022 Advance Notice of Proposed Rulemaking and Request for Public Comment | The FTC has challenged [fake and deceptive reviews], and platforms have sought to combat them, but deception continues to flourish. I agree that these practices are unlawful, and I have supported the FTC's enforcement and guidance in this area. Notably, the Commission recently authorized additional tools [working with states and giving companies an advance warning not to engage in deceptive advertising, which makes imposing fines easier later] to address these issues. Given recent deployment of those tools, as well as ongoing efforts to update our endorsement guides, I do not believe that initiating yet another … rulemaking is the best use of our scarce resources, particularly given the nature of the harm at issue here. And the opportunity cost of launching yet another rulemaking is high, because the division overseeing this rule is also charged with enforcement in the opioids arena. I appreciate that our remedial authority is limited…. But in cases involving deceptive endorsements or fake reviews, there often is no allegation that the product or service did not perform as represented. Moreover, the Commission already has a multi-pronged strategy in place to combat this issue…. Rather than churning out another proposed rule, perhaps we should stay the course on these initiatives and devote the incremental resources to enforcement in other critical areas. The opportunity cost of yet another rulemaking should not be understated. Importantly, as noted above, the division that has responsibility for endorsements also oversees enforcement of the Opioid Addiction Recovery Fraud Prevention Act. Last year, after an 18-month delay not caused by staff, the Commission announced its first case under this statute. For the second consecutive year, deaths from overdoses rose dramatically and now exceed the country's peak deaths from AIDS, car crashes, and guns. Our citizens who suffer from opioid addiction are some of the most vulnerable people in this country; we could use our power and authority to great benefit by devoting more resources to this area. | Go to top Discussion Questions Here are some issues to consider about the proliferation of fake product reviews online: Do you read reviews when making a purchase? What makes you believe they are real or false? Some say reviews made in exchange for a free product are never trustworthy. Do you agree? What are some of the proposed ways to help make reviews more reliable? The role of social media influencers is growing. What do you think is their appeal compared with other customer reviews? What are the downsides? Is government regulation and enforcement the best way to deter fake reviews, or should the focus be on private platform's voluntary decisions and new detection technology? How will AI change the landscape of fake reviews? Go to top Chronology
| | Early 20th Century | As corporate monopolies lead to excessive concentration of power in certain industries, Congress — among other things — passes laws and creates agencies to prevent unfair competition. | 1914 | Lawmakers create the Federal Trade Commission (FTC) to police deceptive and fraudulent corporate advertising practices and to encourage competition. | 1960s–1980s | Early developments in networked computing set the stage for the modern internet that enables e-commerce and, eventually, widespread consumer reviews. | 1969 | ARPANET, the first version of a modern computer network, connects four computers at universities in California and Utah…. Researchers funded by the Defense Department begin working on the precursor to the internet. | 1974 | Computer networking pioneers Vinton Cerf and Robert Kahn coin the term “internet.” | 1978 | The FTC proposes limiting advertising sugary foods to children, prompting charges of overreach that discourages the agency from using rules to enforce antitrust and advertising laws. | 1989 | English computer scientist Tim Berners-Lee creates the World Wide Web, which links pages of online information, greatly expanding the internet's reach and capability. | 1990s-Mid 2010s | Companies follow consumers to the internet, creating an online landscape where even tiny firms have websites; widespread use of online reviews emerges. | 1994 | Entrepreneur Jeff Bezos launches Amazon as an online bookseller that he calls “Earth's biggest bookstore. | 1995 | Entrepreneur Pierre Omidyar creates eBay, which, along with Amazon will trigger radical transformation of retail commerce. | Early 1990s | More than a dozen states enact “food libel laws,” banning negative consumer comments about certain foods, an early attempt to prevent disparaging comments about products. No lawsuits brought under such laws prevail…. Partly in response to such laws, 32 states and the District of Columbia will eventually adopt laws protecting the free speech rights of consumers. | 1997–1998 | A group of Texas cattlemen unsuccessfully sues Oprah Winfrey for negative comments about beef made during a discussion of “mad cow disease” on her TV talk show. | 1999 | First online reviews appear on eBay and a handful of other sites. | 2004 | Harvard student Mark Zuckerberg launches Facebook social media site. | 2009 | Online review site Yelp publishes 4.5 million customer assessments and enables companies to respond to such reviews. | 2010 | Surveys show that nearly all online buyers who read reviews do so at the beginning of the purchasing process | 2012 | Facebook is the leader in online business reviews. | 2015-Present | After online shopping has grown exponentially, online reviews become pervasive, along with deceptive or misleading recommendations…. Regulators try to stem abuses, even as companies try to fight back in the courts and Congress. | 2015 | Footprints Floors in Colorado settles a defamation lawsuit it filed against a customer over a negative 2013 review. | 2016 | In response to the Footprints case, a Colorado lawmaker introduces legislation to protect customers' right to express opinions…. Congress passes the Consumer Review Fairness Act, making it illegal to sue or threaten customers or patients for posting negative reviews. | 2017 | Google is the top site for online business reviews…. Surveys show that 91 percent of millennials trust online reviews as much as they do assessments by friends and family. | 2018 | Tripadvisor has 490 million monthly active users; 95 percent of travelers check online reviews before booking. | 2020-2022 | The FTC proposes stronger guidelines for online product ads and service reviews and boosts policing of fake reviews, including warning social media platforms about inadequate disclosures when reviewers have been given compensation or have other conflicts of interest | 2021 | The Supreme Court curbs the FTC's ability to seek monetary damages (April)…. The agency then uses a different law to notify hundreds of firms that they could face fines for allowing deceptive reviews…. Amazon increasingly kicks firms off its site for violating review rules, including the popular charging-device maker Aukey. | 2022 | The FTC reaches a landmark $4.2 million settlement with Fashion Nova on charges the company blocked negative reviews (January)…. Amazon files a high-profile suit against entities it says operate more than 10,000 groups on Facebook to coordinate paid posters of phony reviews on Amazon…. The FTC announces it will consider new rules to combat phony reviews, potentially mitigating the effect of the 2021 Supreme Court ruling (October)…. Technology firm OpenAI creates artificially intelligent software Chat GPT that mimics human writing, making it even more difficult to spot fake reviews…. The FTC joins seven states to settle for $9.4 million with Google and the radio network iHeartMedia, on claims that Google paid broadcasters to endorse its Pixel 4 cellphone without ever having used it (November). | 2023 | The Supreme Court is expected to decide two cases that could curtail the power of federal agencies to police industries, potentially affecting how aggressively regulators such as the FTC operate (June). | | | Go to top Short Features Whether human or machine, today's would-be-deceivers are more sophisticated than ever at thwarting consumers' efforts to find out the truth about a product or service, according to the Federal Trade Commission (FTC) and other consumer protection groups. Take ChatGPT, a new generation of artificial intelligence that in seconds can write seemingly human-created missives on everything from hammocks to Hamlet. The software is trained on myriad examples of how people speak and write. ChatGPT can create far more believable fake reviews than previous automated options. For example, when asked: “how would a review of a good sushi restaurant read?” ChatGPT responded: “I recently had the pleasure of dining at [Restaurant Name] and was blown away by the quality of the sushi. The fish was incredibly fresh and expertly prepared, with each bite bursting with flavor. The presentation of the dishes was also top-notch, with the chefs taking care to arrange the sushi beautifully on the plate. In addition to the fantastic food, the atmosphere and service at [Restaurant Name] were also excellent. The restaurant had a chic and modern design, and the staff were attentive and knowledgeable about the menu. We were even treated to a demonstration of the chefs making sushi at the bar, which was a nice touch. Overall, I highly recommend [Restaurant Name] for anyone looking for a high-quality sushi dining experience. From the freshness of the fish to the skill of the chefs, this restaurant truly stands out in a crowded field. An internet user generates new text with ChatGPT, a new generation of artificial intelligence that can mimic how people speak and write. One of its many uses is to produce seemingly human-created online reviews. (Getty Images/CFOTO/Future Publishing/Contributor) | And a negative one? “The sushi at this restaurant was absolutely disgusting. The fish tasted old, and the rice was undercooked. I would never eat here again,” or “I was so disappointed with my experience at this sushi restaurant. The service was slow, and the staff was rude. The rolls were small and overpriced. I would not recommend this place to anyone.” Such reviews sound authentic in part because they use specific information, such as the quality of the fish and the knowledgeability of the servers. Fake reviews increasingly include such details to evade detection. Still, many do not, since many impostors aim to write quickly and — most importantly — for multiple sites. A search using a review's sentences could reveal very similar and possibly even identical comments posted elsewhere about different products or services. These similar reviews about disparate items, such as a toaster and vacuum cleaner, are a telltale red flag, according to those who have studied fake reviews, including Fake Review Watch's Kay Dean, consumer advocate Teresa Murray, University of California at Davis researcher Rajvardhan Oak and others interviewed for this article. Besides determining if a review is overly general, a reader can check the reviewer's profile to see if he or she has posted elsewhere. Have they posted suspiciously often? Are the reviews — positive or negative — clustered around the same date? These, too, are red flags, say consumer protection groups. Experts also suggest checking a reviewer's image, if they provide one. Consumers might be surprised to find the same photo being used by multiple reviewers or that the snapshot is of a famous actor (or of someone not so famous that the reader would immediately notice). An image search might reveal that the picture is from a stock photography website. Hannah Rhodes of the consumer-protection oriented Public Interest Research Group says beware of reviews on social media, where so-called “influencers” do not always heed the FTC requirement that reviewers reveal whether they have “a financial, employment, personal or family relationship with a brand.” Nonetheless, about a quarter of Gen Z survey respondents said they trust influencer reviews more than product page reviews, according to a Higher Visibility survey of more than 1,000 U.S. consumers. Has the platform verified that a reviewer made a purchase? Does it sound too good to be true? A “yes” to the first question is a good sign, though not a guarantee, but a “yes” to the second should make the reader wary. Fake-review hunter Kay Dean suggests getting an actual word-of-mouth recommendation for a local service provider, especially for a physician. Subscribing to local email lists might be a place to start. Readers should also question bad reviews. Research suggests that consumers trust bad reviews more than good ones, often failing to consider that some people who post negative comments might be hired by a competitor. However, such cautions are not foolproof. The bottom line, experts say, is shoppers beware. — Kathleen Day
Bibliography
Books
Chiluwa, Innocent, and Sergei A. Samoilenko, Handbook of Research on Deception, Fake News, and Misinformation Online, IGI Global, 2019. Two editors, both professors of communications, discuss misinformation online and fake reviews.
Kumar, Subodha, and Liangfei Qiu, Social Media Analytics and Practical Applications: The Change to the Competition Landscape, CRC Press, 2022. Two business professors address the role of social media in business, including a chapter on social media, fake reviews and machine learning methods.
Novak, Alison N., The New Review Economy: Third-Party Review Sites, Reputation, and Neo-Liberal Public Relations in the Digital Age, Routledge, 2022. A public relations professor examines issues with third-party reviews, including trust and credibility.
Articles
“How to Evaluate Online Reviews,” Federal Trade Commission, October 2022, https://consumer.ftc.gov/articles/how-evaluate-online-reviews. The federal agency that polices deceptive advertising suggests how to spot fake reviews.
Peck, Emily , “How to Spot a Fake Review,” Axios, Aug. 1, 2022, https://www.axios.com/2022/08/01/how-to-spot-a-fake-online-review. A journalist's guide on how to detect — and avoid — fake reviews.
Rhodes, Hannah , “How to Recognize Fake Online Reviews of Products and Services,” The California Public Interest Research Group, March 10, 2022, https://pirg.org/california/edfund/resources/how-to-recognize-fake-online-reviews-of-products-and-services/. A consumer group's suggestions on how to detect fake reviews.
Reports and Studies
“Trade Regulation Rule on the Use of Reviews and Endorsements,” Federal Trade Commission, Jan. 11, 2023, https://www.regulations.gov/document/FTC-2022-0070-0001/comment?sortBy=postedDate&sortDirection=desc. The federal agency that polices fake reviews posts comments from industry, consumer groups and citizens on a recent proposal to beef up rules to curb deceptive endorsements.
He, Sherri, et al., “Detecting Fake-Review Buyers Using Network Structure: Direct Evidence from Amazon,” The Proceedings of the National Academy of Sciences, Nov. 15, 2022, https://www.pnas.org/doi/10.1073/pnas.2211932119. Academic researchers collect evidence that businesses often knowingly encourage the use of fake reviews.
Krishnan, Kavita, and Yun Wan, “The Detection of Fake Reviews in Bestselling Books: Exploration and Findings,” Journal of Electronic Commerce in Organizations, 2021, https://www.igi-global.com/article/the-detection-of-fake-reviews-in-bestselling-books/288313#:~:text=They%20found%20that%20fake%20or,customers%20make%20prudent%20purchase%20decisions. Academic researchers explore how to detect when sales have been manipulated to make a book seem more popular.
Oak, Rajvardhan, and Zubair Shafiq, “The Fault in the Stars: Understanding Underground Incentivized Review Services,” arXiv Computers and Society, Oct. 13, 2022, https://arxiv.org/abs/2102.04217. Academic researchers examine the ways consultants and businesses provide incentives for people to write good reviews.
Stevens, Morgan, and Daniel Castro, “How Lawmakers Can Thwart Fake Reviews,” Center for Data Innovation, Sept. 12, 2022, https://datainnovation.org/2022/09/how-policymakers-can-thwart-the-rise-of-fake-reviews/. A nonprofit think tank that focuses on technology issues and society suggests policy changes lawmakers could make to curtail fake reviews.
Zaman, Mustafeed, et al., “Motives for Posting Fake Reviews: Evidence From a Cross-Cultural Comparison,” Journal of Business Research, January 2023, https://www.sciencedirect.com/science/article/abs/pii/S0148296322008244. These researchers explore the similarities and differences in motivations among shoppers from different cultures — in this case, South Korea and France — who post fake reviews of cosmetics.
Audio and Video
Dean, Kay, “10 Reasons Not to Trust Google Reviews,” Fake Review Watch, January 2023, https://www.youtube.com/watch?v=2UcIN9EgXUo&t=1s. This is one of many YouTube videos by investigator Kay Dean documenting various aspects of the fake-review industry.
“Humanity's Big Bet on Artificial Intelligence,” Tech Policy Press, April 10, 2022, https://techpolicy.press/humanitys-big-bet-on-artificial-intelligence/. This podcast features a conversation between Justin Hendrix, editor of Tech Policy Press, and cognitive scientist Gary Marcus on the benefits of artificial intelligence but also its limits in mitigating misinformation, including deepfakes and AI-written reviews.
Go to top The Next Step AI “Reader Beware: This Gear Review Was Written by an AI Bot,” GearJunkie, Dec. 7, 2022, https://tinyurl.com/45uk3sfa. Editors of an outdoor industry product reviews site used ChatGPT to write a fake product review, hoping to better understand the potential consequences of artificial intelligence for product reviews. Steinmetz, Juergen T., “Fake reviews fully automated,” eTurboNews, Feb. 3, 2023, https://tinyurl.com/yucvr9ws. Fake reviews can undermine a company's reputation and consumer trust in reviews, so greater protection measures and processes to flag and detect bots are needed, according to a travel industry newsletter. Wan, June, “OpenAI's ChatGPT is scary good at my job, but it can't replace me (yet),” ZDNET, Dec. 8, 2022, https://tinyurl.com/mun4rxmr. An IT-focused online publication says ChatGPT is an advanced tool that can be used to generate reviews but fails, so far, to capture the essence of human interaction and communication. Detection Abri, Faranak, “Linguistic Features for Detecting Fake Reviews,” IEEE Xplore, Feb. 23, 2021, https://tinyurl.com/2cwpb3xa. An analysis published by the Institute of Electrical and Electronics Engineers determined that fake reviews use redundant language and long sentences. Guo, Bo, and Zhi-bin Jiang, “What Is the Internet Water Army? A Practical Feature-Based Detection of Large-Scale Fake Reviews,” Mobile Information Systems, Jan. 31, 2023, https://tinyurl.com/5xt5dy7e. Business researchers from Shanghai International Studies University define characteristics that can identify and combat paid reviewers, referred to as members of the Internet Water Army, in this scholarly journal for advances in mobile computing. Plotkina, Daria, Andreas Munzel and Jessie Pallud, “Illusions of truth — Experimental insights into human and algorithmic detections of fake online reviews,” Journal of Business Research, March 2020, https://tinyurl.com/mu7uhz3f. An analysis using subsconsciously used linguistic cues that show the writer is lying to determine fake and truthful reviews demonstrated a need for more sophisticated filtering systems. Laws and Lawsuits Harding, Scharon, “Companies selling fake reviews are “tarnishing” brand, Amazon says,” Ars Technica, Feb. 23, 2022, https://tinyurl.com/mr3dsrcy. The tech news site reports that Amazon filed suit against two companies which, it alleges, sell fake reviews. Marnin, Julia, “Moving company warned customers would pay for bad reviews — ‘$2,500 per day,’ suit says,” The Miami Herald, Jan. 27, 2023, https://tinyurl.com/24skfdpc. A Maine moving business settled with the state attorney general for $125,000, reports The Miami Herald, after the state sued the business owner for violating the federal Consumer Review Fairness Act and state regulations by threatening individuals who left negative reviews. Martínez Otero, J.M., “Fake reviews on online platforms: perspectives from the US, UK and EU legislations,” SN (Spring Nature) Social Sciences, July 21, 2021, https://tinyurl.com/4e8x2rd6. Fake reviews are considered illegal advertising and better regulated in the U.K. than in the United States, a disparity that emphasizes the need for greater protection by the U.S. government agencies. Singer, Paul, et al., “FTC and State AGs Can Continue Joint Case Over Fake Reviews,” Ad Law Access, Feb. 8, 2023, https://tinyurl.com/bdudfv3x. A legal blog says a U.S. District Court held that the Federal Trade Commission (FTC) and states can file joint actions against consumer protection violators, such as companies posting fake reviews. Market and Impact Cramer, Maria, “Five Stars, Zero Clue: Fighting the ‘Scourge’ of Fake Online Reviews,” The New York Times, Jan. 25, 2023, https://tinyurl.com/565ce8cz. Combined with penalties outlined by the FTC, companies such as Yelp and Tripadvisor hope to crack down on the growing threat of companies soliciting fake reviews. Vasani, Sheena, “No, you can't get a 16TB SSD for a hundred bucks,” The Verge, Jan. 16, 2023, https://tinyurl.com/scecv3sb. Sellers of highly rated, fake 16-terabyte, solid-state drives continue to scam buyers through review merging, in which a product listing is edited to make it look as though positive reviews left by consumers for one product apply to its replacement, an unrelated item, according to a technology news website. Wu, Yuanyuan, et al., “Fake online reviews: Literature review, synthesis, and directions for future research,” Decision Support Systems, May 2020, https://tinyurl.com/5x8p4vyv. A science journal focusing the technologies that support decision-making, says a lack of high-quality datasets hinders research on fake online reviews. Go to top Contacts American Civil Liberties Union of Ohio 4506 Chester Ave., Cleveland, OH 44103 614-586-1958 acluohio.org State affiliate of the American Civil Liberties Union that chronicles attempts by business to silence critics and responses by state legislatures. Center for Data Innovation 1101 K St., N.W., Ste. 610, Washington, DC 20005 202-449-1351 datainnovation.org Studies use of data to recommend policies on how technology should be used, including for fake reviews. Consumer Reports 101 Truman Ave., Yonkers, NY 10703 800-333-0663 consumerreports.org Nonprofit consumer advocacy group that independently tests and reviews products and offers advice on how to avoid scams, including fake reviews. Consumer Watchdog 6330 San Vicente Blvd., Ste. 250, Los Angeles, CA 90048 310-392-0522 consumerwatchdog.org Consumer advocate group that focuses on financial issues, including misleading advertising. Federal Trade Commission 600 Pennsylvania Ave., N.W., Washington, DC 20580 202-326-2222 FTC.gov Federal agency that polices unfair, deceptive and fraudulent advertising, including fake reviews. OpenAI 3180 18th St., Ste. 100, San Francisco, CA 94110 650-387-6701 openai.com Creator of ChatGPT, which uses machine learning to mimic human writing. Public Interest Research Group 1543 Wazee St., Ste. 430, Denver, CO, 80202 303-801-0582 pirg.org Consumer advocacy group that provides research and tips on many issues, including fake reviews. Which? 2 Marylebone Rd., London, NW1 4DF, England which.co.uk Consumer research and advocacy group that focuses on, among other things, online reviews. Go to top
Footnotes
Go to top
About the Author
Kathleen Day is an author and a former Washington Post business reporter who is a now full-time lecturer at Johns Hopkins University's Carey Business School, specializing in financial crises and economic history. She previously reported for CQ Researcher on corporate monopolies.
Go to top
Document APA Citation
Day, K. (2023, February 24). Fake online reviews. CQ researcher, 33, 1-19. http://library.cqpress.com/
Document ID: cqresrre2023022400
Document URL: http://library.cqpress.com/cqresearcher/cqresrre2023022400
|
|
|
 |
Feb. 24, 2023 |
Fake Online Reviews |
 |
Jul. 29, 2022 |
Medical Debt in America |
 |
May 07, 2021 |
Reining In Big Tech |
 |
Nov. 27, 2020 |
Protecting Consumer Finances |
 |
Sep. 27, 2019 |
ESG Investing |
 |
Sep. 06, 2019 |
Health and Safety Regulations |
 |
Mar. 16, 2018 |
The Fight Over Net Neutrality |
 |
Mar. 11, 2016 |
Arbitrating Disputes |
 |
Nov. 11, 2011 |
Google's Dominance |
 |
Jan. 23, 2009 |
Regulating Toxic Chemicals |
 |
Dec. 19, 2008 |
Limiting Lawsuits |
 |
Oct. 12, 2007 |
Consumer Safety |
 |
Feb. 17, 1978 |
Consumer Protection: Gains and Setbacks |
 |
Nov. 15, 1972 |
Toy Safety |
 |
Mar. 03, 1960 |
Consumer Protection |
| | | | |
|