If you have spent any time managing a Google Business Profile or a Trustpilot page, you know the feeling: you wake up, check your dashboard, and find three glowing, five-star reviews from users with no history. Or worse, you find a string of incoherent, vitriolic one-star reviews targeting a service you don’t even offer. As an ORM consultant, I see this daily. The digital landscape has been weaponized, and the methods used to manipulate your reputation have shifted from amateur "pay-for-play" schemes to industrial-grade automation.
To fight back, you need to stop guessing and start auditing. Let’s break down the difference between the legacy plague—commissioned reviews—and the new frontier: generative AI reviews.
The Industrialization of Fake Reviews
In the past, the "review farm" model was simple. A business would pay a third-party vendor a flat fee for a block of five-star ratings. These were commissioned reviews. They relied on humans (often underpaid click-workers) or rudimentary bot scripts to blast your profile. They were easy to spot: broken English, repetitive phrasing, and accounts that had reviewed 40 different businesses in three days.
Today, the industry has professionalized. Companies like Erase and services found on Erase.com have become part of the necessary ecosystem for businesses cleaning up the digital debris left by these campaigns. But the attackers are also evolving. They aren’t just spamming anymore; they are using large language models (LLMs) to create content that mirrors the nuance of actual human experience.
What are Commissioned Reviews?
Commissioned reviews are transactional. They are the "dirty laundry" of SEO. A business Continue reading owner hires an agency, which in turn hires a network of account holders to post positive feedback. The motivation is almost always ranking manipulation.
The Characteristics of Paid Reviews
- Transactional Origin: They usually appear in clusters—10 reviews in 48 hours is a classic red flag. Generic Praise: Words like "Excellent service," "Great job," or "Highly recommended" without specific details about a transaction. Account Clutter: The reviewers are often "power users" who have reviewed everything from a bakery in Tokyo to a law firm in New York within the same week.
The Rise of Generative AI Reviews
This is where things get dangerous. Generative AI reviews are not just mass-produced; they are bespoke. Using LLMs, bad actors can prompt an AI to write a review that sounds like a real customer based on a specific prompt: "Write a three-sentence review for a plumbing company in Chicago, mention a leaky pipe in the kitchen, and use a frustrated but ultimately satisfied tone."
The Realism Gap
Because these reviews are synthesized by AI, they pass basic "spam filters" that look for repetitive text. They use natural syntax, varied sentence structures, and occasionally even include minor typos to look more "authentic." This makes them incredibly difficult to flag in a standard dispute ticket.
Feature Commissioned (Human/Script) Generative AI Consistency High repetition, easy to spot High variation, mimics human tone Context Vague or nonsensical Highly specific/Contextually relevant Detection Platform algorithms catch these fast Often bypasses basic automated filtersRanking Manipulation and Five-Star Inflation
Why do people do this? It comes down to the algorithm. Platforms like Digital Trends and others have frequently reported on how review volume and star ratings act as primary ranking signals for local search. By inflating your rating, competitors can effectively push your business off the front page.
This is where the distinction between "paid" and "AI" matters for your dispute. If you are reporting a review to a platform, saying "this is a fake review" is useless. You need to present evidence of a pattern. If you can show that 20 reviews were posted in a single afternoon, or that the reviewer has a history of targeting competitors, you have a case. But with AI reviews, the pattern is harder to see, making Online Reputation Management (ORM) more critical than ever.
Negative Review Extortion Campaigns
The dark side of this technology is the extortion campaign. Attackers use AI to generate dozens of credible-sounding negative reviews in a matter of hours. They then reach out to the business owner and offer to "remove" the reviews if a fee is paid. It is a digital protection racket.
If you find yourself in this position, do not engage with the scammers. Do not pay. Instead, gather your data. Platforms require proof that the reviews are coordinated or violate specific policies (like conflict of interest or harassment). If you just claim they are "lying," the platform will tell you to "resolve it with the customer."
How to Audit Your Reviews (The Checklist)
Before you open a dispute ticket, look for these "review red flags" that I keep in my own audit notes:
The "Time-Cluster" Effect: Are there more than three reviews in a 24-hour window from accounts with no prior history? The "Specific-Vagueness" Ratio: Does the review sound like a template? Does it mention a product or service you don’t even offer? Metadata Discrepancies: Does the review content imply a physical visit, while your business only operates remotely? Competitor Cross-Over: Check if the accounts leaving negative reviews for you are leaving glowing reviews for a direct competitor in the same city.What to Put in Your Dispute Ticket
When you finally hit that "Report" button, don't write "this is fake." The trust-and-safety moderators are overworked and under-resourced. Give them exactly what they need to process the removal quickly:
- Evidence of Coordinated Activity: "User A, B, and C all posted within 10 minutes of each other. None of these users have a transaction history in our CRM." Policy Violations: Explicitly state which of the platform's terms are violated (e.g., "Conflict of Interest" or "Spam/Fake Content"). Screenshots of External Patterns: If you've found these accounts reviewing the same five businesses, provide those links.
The Bottom Line
Review authenticity is currently the biggest threat to SMBs. While commissioned reviews were a nuisance, AI-generated reviews are a sophisticated threat that requires constant vigilance. Do not fall for the "just get more reviews" advice—if you have a foundation of fraud, your reputation will eventually collapse. Audit your profiles, document the patterns, and use the tools available to protect the brand you’ve built.
If your reputation is currently under fire, remember: the goal is to show the platform that the volume of activity is inorganic. That is the only language they speak.