Google AI Overviews Are Mixing Up Small Businesses With Competitors
Entrepreneurs report that automated search summaries conflate their companies with others, surface false complaints, and offer little recourse when errors damage their reputations.

When AI gets it wrong
Damian Mansell searched for his own company and found what appeared to be a crisis. Google's AI Overview for The Plastics Shed, the online building-plastics supplier he launched in early 2025, warned that customer feedback was "overwhelmingly negative." The summary cited delayed deliveries, dishonest staff, and damaged products as "significant recurring issues."
The problem: none of those reviews were actually about his business. The AI had pulled complaints about competitors and companies selling literal plastic sheds. Meanwhile, Mansell was paying Google roughly £700 monthly for advertising that directed users to the same page warning them away.
After weeks of submitting feedback through Google's reporting mechanism, the overview improved. But the episode left Mansell wondering how much business he lost and whether he could recover any of his ad spend. He couldn't.
The new digital storefront
Google's AI Overviews now serve as the first impression for many businesses, replacing the traditional list of links with a single synthesized answer. These summaries draw from websites, Reddit threads, Yelp reviews, decade-old blog posts, and YouTube comments—sources with wildly varying reliability.
Betty Whitney, who runs NW Select Property Management in Idaho, discovered her company was being conflated with a similarly named business that closed years ago. Reviews from property managers in other states appeared under her firm's name. "If I'm a customer looking for a property management company and I look at the AI overview, I wouldn't call me," she said.
Real estate agent Philippa Main relocated from Florida to Northern Virginia three years ago, but Google's AI still confidently states she's been "servicing the Tampa Bay area since 2014"—directly above her Virginia address. Despite combing the internet to find the source, emailing Google, and asking friends to report the error, the false information persists.
Why it matters
This represents a fundamental shift in how businesses must manage their online presence. Search engine optimization professionals describe it as the biggest change to search in nearly two decades. Companies can no longer focus primarily on ranking high in search results—they must now ensure AI systems correctly understand and represent their businesses. For small enterprises competing against larger firms with direct lines to Google representatives, the lack of accountability creates an uneven playing field. The technology processes over five trillion searches annually, meaning even a 90% accuracy rate would produce half a trillion errors.
Limited options for correction
Business owners report few effective remedies when AI summaries go wrong. The standard approach involves repeatedly submitting negative feedback through Google's thumbs-down mechanism and adjusting website content to feed AI crawlers more accurate data. Some hire SEO consultants to increase their "surface area" by publishing more content across more platforms.
Ben Fisher, who runs an SEO consultancy called Steady Demand, found Google's AI warning that his own company was a scam. After investigation, he traced the error to a Reddit thread about a similarly named app. "The big problem is there's nobody to contact," Fisher said. "The other big problem is you search once, and you're done"—without realizing the AI might generate different results the next time.
In a statement, a Google spokesperson said the company's AI experiences "are designed to present a range of perspectives" and are "responsive to people's specific queries." The spokesperson acknowledged that while the "overwhelming majority" of overviews are accurate, there can be cases where they miss context or misinterpret content.
Legal systems are beginning to grapple with accountability. A Canadian musician filed a $1.5 million lawsuit claiming Google's AI falsely identified him as a sex offender. A German court issued a preliminary ruling holding Google liable for false statements in its overviews, though the company plans to appeal.
Mansell's overview eventually became "fantastic," matching his actual positive reviews. But he still pays Google for advertising—without it, people don't find his website. "It does worry me with regard to what can be said about you without any recourse," he said.
These details were first reported by Emily Stewart at Business Insider.
This is an original analysis by the Omega editorial team. Source reporting: AI Watch.
Want systems like this working for your business?
Book a Call