Can Fake Reviews Change Gemini Shopping Recommendations for My Store?

How do reviews feed Gemini shopping answers?

When someone asks Gemini or Google's AI Mode for product recommendations, answers draw on several review sources:

• Product ratings in Google Shopping, supplied by merchants and review aggregators through Merchant Center.
• Store ratings from Google Customer Reviews and approved third-party review sites.
• Business Profile reviews for stores with physical locations.
• Reviews on retailer sites, marketplaces and independent review sites that Google indexes.
• Editorial reviews and comparison articles.

Google says its Shopping Graph holds more than 50 billion product listings, with more than 2 billion updated every hour. Ratings and review text are part of what it uses to describe and compare products. Assistants tend to summarise themes, such as "runs small" or "battery lasts two days", rather than quote star averages alone. Reviews with specific detail therefore matter more than volume.

Can AI systems detect or penalize fake and AI-generated reviews?

Detection happens mostly at the platforms, before reviews reach the assistant:

• Google uses automated systems and human analysts to block and remove fake reviews. When it sees a sudden spike of spam reviews on a Maps profile, it removes them and pauses new reviews. It then alerts the owner and shows a notice explaining why contributions are paused.
• Merchant Center's product rating policies remove reviews with conflicts of interest, such as those written by employees or people with a stake in the product. Merchants must flag spam and disclose incentivised reviews. Incentives cannot depend on the review being positive.
• Other platforms, including major marketplaces and review sites, run their own detection, looking at patterns such as bursts of reviews, linked accounts, reviewer history and repeated wording.

AI-generated review text is increasingly easy to spot in bulk. It tends to be generic, repetitive and free of the concrete details real customers mention. The assistants themselves are not a reliable judge of review authenticity. They mostly rely on what the platforms keep. But because they summarise themes, a burst of vague praise adds little to what they say.

What happens if a store is caught using fake reviews?

• Reviews are removed, so any rating gain disappears.
• Profiles can carry warnings. Under undertakings agreed with the UK's Competition and Markets Authority in January 2025, Google committed to put warning alerts on the profiles of UK businesses that boost their ratings with fake reviews, and to restrict new reviews.
• Accounts can be restricted. Google can restrict Business Profiles for fake engagement, and Merchant Center can suspend accounts that break its policies.
• Legal penalties. The US Federal Trade Commission's rule, in force since October 2024, bans fake and AI-generated reviews and allows civil penalties. The UK banned fake reviews under the Digital Markets, Competition and Consumers Act from April 2025, with fines of up to 10% of global turnover. EU consumer law has banned fake reviews since 2022.
• AI answers may repeat the damage. Warning notices, news coverage and complaints about fake reviews become part of what assistants find about the store.

Can a competitor's fake reviews hurt my store?

They can, briefly, and extortion attempts are common: scammers post fake one-star reviews and demand payment to remove them. Google has said it is upgrading its systems to stop such schemes faster. If it happens to you:

1. Do not pay and do not respond with fake positive reviews.
2. Report the reviews through your Business Profile, Merchant Center or the relevant platform, citing the policy broken.
3. Document patterns: dates, wording, reviewer accounts and any demands.
4. Reply publicly and calmly to reviews that are visible, stating facts without accusations you cannot prove.
5. Keep collecting genuine reviews, which dilute and outlast the fake ones.

What actually improves how Gemini describes your store and products?

• Ask every buyer for a review, a few days after delivery, with nothing offered in return and without suggesting what to write.
• Ask every buyer. Detailed reviews help most, because assistants summarise recurring themes, but you cannot steer what customers write, so ask every buyer for an honest review in their own words.
• Use Google Customer Reviews or an approved review aggregator so store and product ratings reach Google.
• Keep product data accurate in Merchant Center, since assistants combine reviews with specifications and prices.
• Respond to negative reviews with fixes, and fix the underlying problems.
• Earn independent reviews from publications and creators who test products.

Why assistants sometimes treat small stores as risky is covered in why Gemini might treat smaller Shopify stores as scams. How to check what Gemini currently says is covered in how to check what Gemini says about your business.

What about reviews on your own website?

Reviews on your own product pages also feed AI answers, because assistants read them along with the rest of the page. The same rules apply. Show reviews from real buyers, including critical ones. The FTC rule also bans suppressing negative reviews through threats or false accusations. Use a review platform that verifies purchases and labels verified buyers. If you add review structured data, it must describe reviews visible on the page. Google's review snippet guidelines also exclude self-serving reviews of a business on its own site from local business and organisation rich results. Product reviews on product pages remain eligible. A page showing only perfect scores looks less credible to shoppers and gives assistants less useful detail than an honest mix.

What might an illustrative case look like?

Consider an illustrative online homewares store that buys 200 five-star reviews from a service to lift its product ratings before the holiday season. Within weeks, most are removed, and its Business Profile shows a notice that reviews have been paused. The store's real reviews, which had mentioned slow delivery, now dominate what remains, and Gemini's summaries mention delivery delays. The store stops the scheme, fixes its courier arrangements and starts asking every buyer for an honest review in their own words a few days after delivery. Over the next six months, several hundred genuine reviews build up. Gemini's summaries begin to mention careful packaging and accurate product descriptions instead.

What do people ask about fake reviews and AI shopping?

Do AI-written reviews from real customers count as fake?

Reviews must reflect a genuine experience. A customer using AI to help phrase their own honest review is different from generated reviews with no real experience behind them. Check each platform's rules.

Can I offer a discount for a review?

Don't. Google Business Profile bans incentives outright, Merchant Center requires any incentivised review to be flagged as such, in the US an incentive tied to a positive review is illegal, and in the UK and EU incentivised reviews must not be concealed. Asking every customer for an honest review, with nothing in return, avoids all of these.

How quickly do removed fake reviews stop affecting AI answers?

Once removed from the source and recrawled, usually within weeks. Answers built on older summaries can lag.

Will one bad review hurt my Gemini recommendations?

Rarely. Assistants summarise patterns across many reviews. A steady flow of genuine reviews outweighs isolated complaints.