AI Google Reviews: Where Help Ends and Fakery Begins
AI Google reviews are fine when a real customer uses a tool to say their own experience and posts it themselves. They are fake when the experience, the author or the volume is invented.
AI Google reviews fall on one of two sides of a line. A review generated for an experience that never happened, or written by a business and posted as a customer, is fake and prohibited. A real customer who uses speech-to-text or a grammar tool to express their own experience, then reads, edits and posts it themselves, is writing a genuine review with a tool. The test is who had the experience and who controls the words and the post button, not whether software touched the text.
This post covers what Google and the US Federal Trade Commission say, a four-question checklist for judging any AI review tool, and how AI fits when the business is the one speaking, in owner replies.
AI Google reviews: what Google's policy says
Google's Maps Prohibited and restricted content policy does not mention AI by name. What it does say applies to AI as much as to anything else:
- Contributions should reflect a genuine experience at a place or business.
- Fake engagement, including content not based on a real experience, is not allowed and will be removed.
- Merchants may not solicit or encourage content that does not represent a genuine experience.
- Content posted from multiple accounts by, or at the request of, one person is fake engagement.
- Reviews or ratings that have been paid for, directly or in kind, are fake engagement.
- Content based on a conflict of interest, such as current employment or a personal relationship with the business, is not allowed.
- Merchants may not request that specific content be included in a review.
Read together, those lines make the question about the author and the experience, not the tool. Spell-check is software. Dictation is software. A language model that invents a visit is also software, and it breaks the first rule.
What the FTC rule says about AI-generated reviews
In the United States, the Federal Trade Commission's rule on consumer reviews and testimonials took effect in October 2024. The FTC's announcement of the final rule says it bans fake or false consumer reviews, including AI-generated fake reviews, and testimonials that misrepresent that they are by someone who does not exist or who did not have actual experience with the business or its products. It also bars businesses from buying or selling reviews and lets the FTC seek civil penalties against knowing violators.
The FTC's Questions and Answers page adds that a business can be liable for writing fake reviews itself, or for buying or distributing reviews it knew or should have known were fake. It also says a business should not provide the text for a testimonial without a reasonable basis.
This is general information, not legal advice. If you operate in the US or sell to US customers, read the rule or ask a lawyer about your situation.
What is not allowed
Three patterns are on the wrong side of both Google's policy and the FTC rule.
- Reviews for an experience that did not happen. Prompting a model with "write a five-star review for a plumber in Leeds" and posting it. There is no customer and no visit.
- Reviews written by the business and posted as customers. An owner or staff member writes the text, or has an assistant write it, and posts it from a personal or spare account. This is also a conflict of interest.
- Mass-generated text. Dozens of reviews with similar phrasing, posted from several accounts or handed to people to paste. Even where each poster did buy something, the words are not theirs, and the pattern is the kind Google describes as unusual volumes of reviews.
Dressing any of these up as a "review generator" does not change what they are.
What is reasonable
A real customer, with a real experience, uses tools to say it more easily. Examples that sit comfortably inside the rules:
- A customer speaks into their phone and speech-to-text turns the words into text. The ideas are theirs. The tool is a keyboard replacement.
- A customer runs a grammar or phrasing check on their own draft, then accepts, rejects or rewrites the suggestions.
- A customer whose first language is not English writes in their own language or in rough English and uses a tool to tidy it.
The pattern in each case: the person decides what happened, what to say and what rating to give, can change every word, and posts it from their own Google account. If you want a hands-on look at the speech side, see how to leave a Google review by voice.
A four-question checklist for any AI review tool
Use these on any "AI review generator", including ours. If a tool fails any one, treat it as unsafe for your business.
- Whose experience is it? The review must describe something the person who posts it actually experienced. A tool that can produce a review without a real customer behind it fails here.
- Who decides the rating and the content? The customer. A tool, a staff member or the business must not pick the star rating or the story, and the business must not ask for specific content.
- Who presses post? The customer, from their own Google account. If the business, the tool or an agency posts on someone's behalf, it is not their review.
- Can the customer change every word, or discard all of it? If the text is locked, auto-submitted or the only option offered, the tool is deciding, not helping.
A fifth check applies to the business using the tool: does everyone get the same ask, or do you filter who sees it? Selectively soliciting only positive reviews is prohibited whatever tool carries the message. See what review gating is.
How GMBQrcode's review flow measures up
GMBQrcode's review page works like this. A customer scans the business's QR code, or taps the NFC stand. On the page that opens, they can say what they thought out loud using the microphone button, which uses the phone browser's built-in speech-to-text to turn their words into text, or they can type. There is an optional "Refine with AI" button that tidies grammar and phrasing. The customer can ignore it, or edit the result. Then the customer copies the text and posts it on the business's Google review page themselves, from their own Google account.
Held to the same four questions:
- Whose experience? The customer's. The text begins with what they said or typed.
- Who decides rating and content? The customer. The Refine button reworks the wording of what they gave it, and the customer can change or ignore the result.
- Who presses post? The customer, on Google's own review page, from their own account. Nothing is posted automatically.
- Can they change every word? Yes. The text sits in an editable box, and nothing forces them to use the AI version.
The aim is narrow. It removes effort for people who are willing to review but would not bother to type one. It does not post anything for the customer.
One thing to know before you rely on that answer. The review page can also offer a suggested draft based on the services the customer selects. Hold it to the same test. A suggested draft is only acceptable if the customer reads it, changes it until it describes what actually happened to them, and decides for themselves whether to post it. If you want the strictest reading of the rules, tell customers to speak or type their own words and treat the draft as optional.
The rules from Google still apply to you as the business: ask everyone, offer no incentives, and do not script what customers say. If you want to try the review flow, look at the features page or sign up free.
AI for owner replies is a different case
Replies are the business speaking for itself, so there is no experience to fake. Using AI to draft a reply to a review is closer to using a template than to writing a review.
The rules that matter here are different:
- A reply must still follow Google's content policies, and Google checks replies before they appear. Google's help page on managing reviews says this usually takes up to 10 minutes but can sometimes take up to 30 days.
- Replies are public and the reviewer is notified, so a reply that reads as generic or defensive will be noticed.
- Google's tips for reviews advise keeping replies short, professional and helpful, and not sending the same thank-you to everyone.
- Never include a reviewer's private details, and never answer with an offer to change or remove the review in return for a reward.
Use AI for the first draft, read it, and add what only you know: what happened, what you will change, how to reach you. The free AI review reply generator is a starting point for that. For the wider approach, see how to respond to negative Google reviews.
FAQ
Can you use AI to write a Google review?
A real customer can use AI to help express their own experience, for example dictating by voice or tidying grammar, provided they read it, edit it and post it from their own account. Using AI to invent a review for a visit that did not happen is fake engagement. The deciding factor is whether the experience and the decision to post belong to the customer.
Does Google ban AI-generated reviews?
Google's Maps policy does not mention AI by name. It bans reviews that are not based on a real experience, reviews posted from multiple accounts by one person, paid reviews and conflict-of-interest content. An AI-written review for an experience nobody had falls under those rules.
Is it legal to use an AI review generator for my business?
In the US, the FTC rule bans fake reviews, including AI-generated ones that misrepresent a reviewer who does not exist or has no real experience. A tool that helps real customers phrase their own reviews is a different thing from a tool that manufactures reviews. This is general information, not legal advice.
Can I use AI to reply to Google reviews?
Yes. Replies come from the business, so AI is a drafting aid. Check the draft against Google's content policies, make it specific to the reviewer's feedback, then post it under your Business Profile.
How do I tell an AI-written fake review from a real one?
No single sign is proof. Look at the reviewer's other activity, repeated phrasing across several reviews, and details that do not match anything you sold. Report a review that you believe breaks policy, and see spotting fake Google reviews.
Check your own setup first
Before you adopt any AI review tool, run it through the four questions above. If you want to see how the review flow works from the customer's side, start with the free Google review QR code generator, then compare plans on the pricing page.