Does Google Warn Against Buying Brand Mentions to Influence AI Recommendations?

What exactly has Google said?

The guide's full passage is short. It says Google's generative AI features "can show what's being said about products and services across the web, including in blogs, videos, and forum discussions." It then warns that seeking inauthentic mentions is less helpful than it seems, for the reasons above.

What counts as an inauthentic mention?

Google does not define the term. A reasonable reading is a mention that exists because you paid for it or planted it to influence rankings or AI answers, rather than because an independent person chose to write it. That covers:

• Paid inclusion in "best of" lists or reviews that is not disclosed as advertising.
• Posts by staff or hired accounts in forums, comment threads and Q&A sites, presented as ordinary users.
• Networks of sites set up to mention your brand, covered in whether buying extra domains and building microsites increases LLM mentions.
• Mass-produced articles placed on other sites purely to repeat your brand name next to category terms.
• Hidden instructions in pages or links meant to steer assistants, the tactic Microsoft describes in its AI recommendation poisoning research.

The test is not whether money changed hands but whether the mention pretends to be independent when it is not.

What stays legitimate?

Plenty of paid and organised activity is fine, as long as it is honest:

• Clearly labelled sponsored content, accepting that Google may give it little weight. Paid links inside it should carry rel="sponsored".
• Affiliate programmes with disclosed relationships, where you pay for referrals, not verdicts. See how to build an affiliate program that earns product mentions in AI answers.
• Digital PR, where journalists choose to cover your story or data, covered in whether digital PR improves SEO and AI search visibility.
• Staff taking part openly in community discussions, saying who they work for.
• Customer reviews requested honestly, from every customer, with no incentive and no scripted wording.
• Pitching editors on merit for list inclusion, described in how to get your brand added to third-party listicles.

Why does Google compare it to paid links?

Because the pattern is the same. When links drove rankings, a market grew for buying them, and Google responded with its link spam policy and algorithmic detection. In its December 2022 link spam update, Google said its SpamBrain system could detect both sites buying links and sites used to pass outgoing links. Links found this way are ignored rather than counted. Illyes's comparison suggests Google expects to treat bought mentions the same way: discount them, so the money is wasted.

Two points are worth noting. First, Google's guide does not threaten penalties for mentions; it says they are less useful than they seem. Second, Illyes reportedly said Google was unsure how useful ordinary mentions on the web actually are. So paying for mentions is spending on a signal whose value Google itself will not confirm, in a form Google says it tries to ignore.

Does the warning apply to ChatGPT, Perplexity and Copilot?

Google's statement covers Google's own AI features. Other assistants set their own rules:

• Microsoft Bing rewrote its webmaster guidelines in February 2026 to cover Copilot answers, with a section on prompt injection and AI manipulation that can lead to demotion or delisting.
• OpenAI says ads in ChatGPT do not influence answers, but it does not publish a policy on paid mentions in the sources ChatGPT retrieves.
• Perplexity and ChatGPT retrieve pages from the web in real time, so a paid article that is indexed and open to crawlers can be read like any other page.

That last point is why vendors claim bought mentions work. We have not tested whether they change AI answers, and we know of no public study that measures the effect over time. What is clear is the downside. If the publisher keeps its sponsored section out of search indexes, assistants that search may never see the mention. A visible "sponsored" label is part of the text an assistant reads. And a tactic Google and Microsoft both call manipulation is one that can stop working without notice. How paid and earned sources compare is covered in whether organic AI citations carry more trust than paid placements.

What are the legal and reputational risks?

Search engine policy is not the only constraint. In the US, the Federal Trade Commission's Endorsement Guides, revised in 2023, require material connections, such as payment or free products, to be disclosed. Undisclosed paid mentions presented as independent opinion can be deceptive advertising. Posts by staff or hired accounts pretending to be customers carry the same risk, and when they are exposed, the story itself becomes the most visible mention of your brand.

There is also a practical risk. Mentions bought in bulk tend to sit on low-quality sites, next to other paid placements. If assistants do pick them up, they associate your brand with those sites rather than with the publications buyers trust.

How should you evaluate a vendor selling AI mentions?

Some AI visibility tools and agencies now sell placements on the sites AI answers cite. Before paying, ask:

1. Where will the mention appear, and can you see the exact pages in advance?
2. Is it disclosed as sponsored, and are paid links marked rel="sponsored"?
3. Is the site editorially independent, or does it exist mainly to sell placements?
4. Who writes the content, and is it produced at scale by AI?
5. What is promised? Google's June 2026 guidance notes that no tool or service can guarantee performance.

If the honest answers are "undisclosed", "a network of similar sites" or "guaranteed AI rankings", the offer is the kind of mention Google has warned about.

What should you do instead?

The mentions that last are the ones people choose to make. They come from being worth citing:

• Publish something others will reference, such as original data, pricing studies or a clear explanation nobody else has written.
• Make your own pages the best source for your product facts, so assistants have an accurate page to use.
• Earn coverage through PR and relationships with the publications that shape your category.
• Track which sources AI answers cite for your key questions, and focus effort there.

The difference between linked and unlinked mentions is covered in whether backlinks help GEO and whether unlinked mentions matter.

What might this look like in practice?

Consider an illustrative payroll software company, Ulvoway, offered a package of 40 "AI-optimised brand mentions" for $6,000 a month. The sample pages were on general business blogs with no bylines, each listing "top payroll tools" with Ulvoway first and no sponsored label. The marketing lead turned it down, citing Google's guidance and the FTC disclosure rules. Instead, the company published a yearly study of payroll errors by company size, using anonymised data from its own customers. Two trade publications and a national business site covered the findings, and within two months AI answers about payroll accuracy began citing the study and naming Ulvoway.

What do people ask about buying brand mentions for AI?

Does Google penalise sites for buying mentions?

Google's guide says inauthentic mentions are less helpful than they seem. Illyes compared them to paid links, which Google ignores. Neither describes a specific penalty.

Is sponsored content against Google's rules?

No, if it is clearly disclosed and paid links are marked rel="sponsored" or rel="nofollow".

Do authentic mentions help in Google's AI features?

Google's guide says its AI features can show what is said across the web. Illyes reportedly would not confirm a direct benefit.

Can bought mentions still influence ChatGPT?

A paid page that assistants can retrieve may be read. No public study shows a lasting effect, and the disclosure and reputational risks remain.