1. The Quiet Crisis Behind the Star Ratings
Imagine you’re scrolling through a product page late at night, trying to decide between two nearly identical items. One has a 4.8-star rating with glowing reviews that describe life-changing ease. The other has a 4.2-star rating with a handful of passionate complaints. Which do you choose? For most of us, the reviews make the difference. Online reviews have become the modern equivalent of word-of-mouth advice—a digital village square where strangers help us decide what to eat, where to sleep, and what to buy for the people we love. But what happens when those voices aren’t genuine? What if some of those five-star testimonies were bought, fabricated, or generated by artificial intelligence? New research from the Coalition for Trusted Reviews (CfTR) suggests that this worry is no longer fringe. In fact, it is reshaping how we shop. The CfTR is a remarkable alliance, bringing together companies that are otherwise fierce marketplace rivals: Amazon, Booking.com, Expedia Group, G2, Holidaycheck, Tripadvisor, and Trustpilot. Their common cause is the integrity of reviews—the shared trust that makes their platforms usable. The coalition surveyed 6,000 consumers across the United Kingdom, Italy, and the United States, and the findings are both familiar and unsettling. Eight in ten people in the UK (85%), the US (82%), and Italy (80%) say they use online reviews at least sometimes when weighing a purchase. That means reviews are not a niche feature of e-commerce; they are the backbone of modern consumer decision-making. Yet at the same moment, a vast number of us are losing faith in what we read. Some 74% of UK consumers, 73% of Americans, and 61% of Italians say they have consciously walked away from a purchase because they feared the reviews were fake or misleading. Think about what that means: the system designed to help us buy with confidence is increasingly making us hesitate. It is a quiet crisis, hidden in plain sight, and it touches nearly every corner of our online lives.
2. The Real Cost of Fake Reviews
When we choose not to buy because we don’t trust the reviews, the consequences go far beyond a single abandoned shopping cart. For consumers, there is the lost time spent researching, the lingering uncertainty about whether we made the right call, and the nagging feeling that the internet is becoming a place where nothing can be taken at face value. For honest businesses, the damage is even more painful. A small café that has worked hard to earn genuine five-star reviews can be overshadowed by a competitor with purchased praise. A family-run hotel that patiently answers every negative comment can slip down the rankings because of a flood of fabricated testimonials. As Becky Foley, Executive Committee member of the Coalition for Trusted Reviews and VP, Head of Trust & Safety at TripAdvisor, puts it: “When reviews are fake or manipulated, the consequences go far beyond one misleading rating. Consumers risk making decisions based on information that isn’t genuine, while businesses that play by the rules face an uneven playing field. Over time, this undermines trust across the online marketplace.” The problem is not static; it is getting more sophisticated by the month. The research reveals that online platforms already invest heavily in fighting misuse, but review fraud has evolved into a cross-border, cross-platform enterprise. Bad actors—people who deliberately try to break platform guidelines and mislead consumers—do not confine themselves to one website. They reuse identities, tactics, and networks across different jurisdictions. One group might create fake reviews for restaurants in Rome, gadget brands in London, and holiday rentals in Florida, all using similar patterns that are difficult for any single company to detect. This is why the CfTR argues that no single platform or regulator can disrupt fraudulent activity on its own. A review that is caught and removed on one site may have already influenced hundreds of buyers on another. The fraudsters are agile, organized, and constantly adapting. Meanwhile, the rest of us are left trying to read between the lines, wondering whether the person behind that glowing review was a real customer with a genuine experience, or a paid actor in a carefully orchestrated campaign.
3. The Verification Dilemma
One of the most obvious solutions to fake reviews is to verify the people leaving them. If a platform could confirm that a reviewer actually bought the product, stayed at the hotel, or used the service, then harmful fake reviews should become much harder to create. Many platforms already use review and user verification as part of their arsenal, and it can be surprisingly effective. But the same research that shows how much we fear fake reviews also reveals a deep tension in our own expectations. We want authenticity, but we are not always willing to do the work to prove our own authenticity. In the UK, 56% of consumers say that being required to provide ID would make them less likely to leave a review. The figures are also significant in the US (46%) and Italy (43%). That is a genuinely uncomfortable finding. On one hand, we demand protection against fraud; on the other, we are reluctant to share the kind of identifying information that would offer that protection. It is easy to understand why. Nobody wants to upload a passport or driver’s license just to tell the world that a folding chair was exactly as described. There are legitimate concerns about privacy, data security, and the feeling that we shouldn’t have to prove our honesty just to leave a comment about a hotel room or a laptop. Yet without some form of verification, the systems remain vulnerable to manipulation. This is the classic trade-off that digital platforms face every single day: how to make reviews more trustworthy without making them feel invasive. The research suggests there is no simple answer. Perhaps the solution lies in offering a range of verification options, some lighter than others, so that consumers can choose a level that feels comfortable. But solving this dilemma will require more than technical wizardry. It will require honest conversation about what we are willing to give up in exchange for greater confidence in the information we rely on.
4. The Complicated Arrival of AI
Just when we thought the fake review problem was complicated enough, along came artificial intelligence. The research found that 70% of UK consumers would trust a review less if they knew it had been generated or written with AI assistance, alongside 58% in Italy and 51% in the US. At first glance, that seems like a reasonable response. We want reviews written by real people, with real experiences, real emotions, and all the small, imperfect details that make a story credible. A computer-generated review might be grammatically perfect, but it lacks the quirks that signal authenticity. But the relationship between AI and reviews is not quite so black and white. The CfTR is careful to point out that the use of AI does not automatically make a review untrustworthy. There are legitimate uses, such as helping a non-native speaker translate their thoughts into a different language, or summarizing a genuinely long experience into a concise format. Each online platform has its own stance on whether AI assistance aligns with its guidelines. Some may allow it explicitly, others may ban it outright, and many are still trying to decide. The real danger is not the mere presence of AI, but the intention behind it. A human being who writes a glowing review of their own business is still lying even if they type it out themselves. An AI tool that generates hundreds of five-star reviews for a product it has never seen is a fraud on an industrial scale. What makes AI truly difficult is that it can be used to produce convincing, human-like text at a speed no person could match. It can learn the idioms, rhythms, and emotional cues that make readers feel safe. That capability can be used for good or ill, which is why the coalition insists that technology alone cannot tackle the problem. Continued investment in trust and safety teams is essential, combining the best of machine learning with the human judgment needed to identify emerging threats, interpret complex fraud signals, and ensure that enforcement is both effective and fair. A machine might flag a suspicious pattern, but it takes a human to understand context, nuance, and intent.
5. No Company Can Do It Alone
If there is one message that echoes throughout this research, it is the simple but profound truth that review fraud is not a problem any single company can solve in isolation. The networks behind fake reviews are not geographically contained, and they are not loyal to any one platform. They move wherever the money is, adapting their methods to whatever countermeasures are introduced. That is why the Coalition for Trusted Reviews is calling for coordinated action from industry, governments, and regulators. The goal is not to replace the efforts of individual platforms, but to connect them. By sharing intelligence responsibly, companies can build a clearer picture of fraud across the entire ecosystem. They can spot a suspicious email address that has been used to create accounts on five different review sites. They can identify a group of reviewers who all seem to travel together, but never actually visit the same venues. They can recognize patterns that are invisible to any one company, but glaringly obvious when the data is combined. None of this is easy. There are legal barriers, competitive tensions, and legitimate concerns around data protection and privacy. But the alternative is to keep fighting a global game of whack-a-mole, where bad actors are constantly a step ahead. Becky Foley again: “As bad actors adopt increasingly sophisticated tactics, no single company can tackle review fraud alone. That is why the coalition is calling on industry, governments and regulators to take coordinated action—sharing intelligence responsibly, building a clearer picture of fraud across the ecosystem and closing the gaps between platforms and markets that fraudsters exploit.” For consumers, this is a hopeful message. It means the people who run the platforms we use every day are not ignoring the problem. They are talking to each other, sharing what they know, and looking for ways to make the entire ecosystem safer. But it also means that progress will take time. Trust is not restored overnight. It is rebuilt, review by review, through consistent and visible action.
6. Looking Ahead with Hope and Innovation
The findings from the Coalition for Trusted Reviews arrive at a moment when trust itself has become a precious commodity. In an increasingly digital, often impersonal commerce world, consumers are hungry for honest signals. We want to know that the product we buy will actually work, that the room we book will actually exist, and that the restaurant we choose will actually welcome us. Online reviews, at their best, provide that reassurance. At their worst, they contribute to a growing sense that the internet is filled with illusions. For companies and technologists working in retail and commerce, this research is both a warning and an opportunity. The warning is clear: if platforms do not act decisively, they risk losing the very trust that makes their models work. The opportunity is equally clear: there is a real competitive advantage in being known as the platform that users can trust. This is why innovation in trust and safety is becoming just as important as innovation in product design, delivery speed, or pricing. The 2026 RTIH Innovation Awards, which celebrate global retail technology innovation in a fast-moving omnichannel world, will likely see trust, authenticity, and fraud prevention take center stage. The ceremony, to be held at The HAC in Central London on Wednesday, 4th November, will bring together the industry’s brightest minds to recognize those who are shaping the future of retail. But beyond the awards and the celebrations, the deeper challenge remains: how do we keep the human element alive in a digital marketplace? How do we ensure that the voices guiding our decisions are real, honest, and free from manipulation? The research from the CfTR suggests that consumers, businesses, and platforms all share a common interest in answering those questions. No one wants to go back to a world of empty words and hidden agendas. With continued investment in human trust and safety teams, responsible use of AI, sensible verification methods, and cross-industry collaboration, there is reason to believe that online reviews can be both truthful and useful again. It will not be easy, but it is one of the most important debates in modern commerce. After all, every purchase includes a leap of faith. The question is whether we are willing to build a system that makes that leap a little safer.