In February 2025, Google had a “glitch.” That’s what they called it, at least. Thousands of business owners across the globe logged into their Google Business Profiles and saw something terrifying: their review counts had dropped like a bad habit. Not a few here and there. In some cases, dozens. Hundreds. A digital vanishing act worthy of David Copperfield.

Google, as usual, said “we’re aware of the issue” and “we’re working on a fix.” And by February 11th, things slowly started to return to “normal.” But if you think this was just a random hiccup in the Matrix, I have a theory for you.

And yes — it’s 100% a theory. I have no evidence. Zero confirmation. Just pure belief. But it’s the kind of belief I’d die for.

Here it is:

I believe Google let Gemini audit reviews to see which ones it would delete… but instead of just flagging them, Gemini actually dropped the review count.
And then Google panicked.


So What Actually Happened?

Let’s recap the weirdness.

  • Businesses noticed that their review counts didn’t match the actual number of reviews.
  • In many cases, the reviews were still visible when you clicked into them — but the total count displayed on the profile was wrong.
  • Google called it a “known issue” and said they were “rolling out a fix.”
  • Forums like Reddit lit up with business owners comparing stories, panicking, trying to understand why Google was gaslighting their hard-earned stars.

Google gave us a few reasons:

  • “Reviews might be missing because they were spam.”
  • “Maybe they violated our policies.”
  • “Could be from before your business opened.”
  • Yada yada.

But here’s the part nobody questioned: they never explained why the counts dropped while the reviews were still visible.

That’s not a spam issue. That’s not a policy violation. That’s a misfire in the system. Or, more likely… an AI audit that went rogue.

Google told Gemini to audit your reviews. Gemini deleted the count instead. And now Google’s pretending it was just a glitch.

Enter: Gemini — Google’s AI Review Bouncer

Google officially announced on April 7th, 2025 that they’re now using AI to detect fake reviews and review bombing on Google Business Profiles. Their tool of choice? Gemini.

But Gemini didn’t just fall out of the sky that day.

Gemini 2.0 launched for consumer use back in December 2024. That gave Google a few months to integrate it internally, train it on new tasks, and — hypothetically — start letting it peek behind the curtain of Maps and Business Profiles.

You mean to tell me that it’s just a coincidence Gemini got involved in reviews after the February glitch? Please.

You can’t convince me Google wasn’t quietly testing Gemini on real-world data to see how it handled spammy reviews. The problem? It didn’t just report what it found — it started messing with the actual public-facing review counts.

Like giving a toddler scissors and telling them to “just pretend to cut the hair.” You know how that ends.


Why Google Won’t Admit It

Let’s get one thing straight: Google will never confirm this. Ever.

Why? Because the narrative that AI acted on its own is terrifying. It feeds every Skynet-obsessed, Terminator-fearing, tech-apocalypse conspiracy out there. And Google knows it.

Imagine the backlash if they said:

“Oops! Our AI misunderstood the assignment and accidentally wiped out parts of your business’s online credibility. But it’s fixed now, probably.”

Instead, they called it a glitch. Rolled out a fix. Hoped we’d forget. And most people probably did.

But not me. I see you, Gemini.


The AI That Cried Spam — and Might Brand You For It

Here’s the real danger: Gemini didn’t just misbehave — it over-applied Google’s already murky review policies.

Spam? Fake? Inappropriate? Those labels are fuzzy even when a human is making the call. Now imagine an AI trained to be overly cautious. It sees a review that looks too enthusiastic, too vague, or too perfectly timed after a competitor gets dinged — boom, flagged. Count deducted. No questions asked.

The reviews themselves? Still there. But the count drops. And to customers scrolling by your profile, that drop reads like something’s off.

But it gets worse…

If Gemini decides too many of your reviews are “fake,” your business could end up with a public-facing alert — a digital scarlet letter — slapped right on your Google Business Profile.

Straight from Google’s own review policy:

“If we detect a large number of policy-violating reviews for a place or business, we may add a label or warning to let users know.”
Google’s Review Policy

That’s right — not only could Gemini quietly ghost your reviews, it could also call you out publicly like a hall monitor on a power trip. Whether the reviews were legitimate or not doesn’t matter. Gemini is judge, jury, and brand executioner.


Why You Should Care

Look — this isn’t just about a past glitch. It’s a peek into the future of how your business’s reputation will be judged.

AI is now the bouncer at the door of your Google reviews.

You can have a loyal customer base, real feedback, and a five-star average — but if Gemini decides one of those reviews smells fishy, it might quietly shadow-ban it from your profile.

And you’ll never know why. Or how. Or what to do about it.


Final Word

I’m not saying it was Gemini.

Actually, yes I am.

It was Gemini. That’s my story and I’m sticking to it. There’s no proof. Just a glitch, a timeline, a machine learning model with a little too much confidence, and a suspiciously delayed announcement.

I believe it.

And you should at least wonder about it.