What happened

Articles on company websites often have an author listed underneath: a name, a headshot and a line about their credentials. Google’s documentation for site owners states that if that author is fabricated, for example with an AI-generated headshot, a made-up name or false credentials, to make the content look as if a human expert wrote it, that is a form of deception.

The document goes on to say that this kind of deception makes a page untrustworthy to both readers and the automated systems Google uses to judge page quality, and that Google treats it as a sign of a low-quality page. The document doesn’t say outright whether this costs a page its rankings. But it opens by saying Google wants to put helpful, reliable content first. Following that logic, a page treated as untrustworthy would likely struggle to rank.

So should you drop author names altogether? No. Google actually strongly encourages adding accurate author information; what it cares about is whether the author is real.

Google has no objection to content written with AI. Back in 2023 it said its focus is on the quality of content, rather than how content is produced. So using AI to write isn’t the problem in itself; pretending a particular expert wrote it is.

Why it matters to you

What Google is calling out may well be on your own website. Say you run an online shop in Tainan selling health supplements, and the articles on your site are written by an outside agency. They use AI to turn out twenty articles a month, each one signed “Dietitian Lin”, with a warm, smiling headshot and a bio that says “ten years of clinical experience”. But nobody by that name works for you: the photo was made with AI, and the ten years of experience were made up by the writer. You went along with it because having a dietitian’s name on the articles looked more professional.

By the wording of Google’s document, this example ticks all three boxes: an AI-made headshot, a made-up name and false credentials. The professional look you wanted to borrow gets you “this page can’t be trusted” instead. It’s the same with customers: if someone wants to ask Dietitian Lin a question, or searches for that name, and finds nothing at all, their trust in your whole site takes a hit.

Google’s document is about Search and doesn’t mention AI assistants such as ChatGPT, Gemini or Perplexity. But customers now also ask these AIs directly which business is best, and when they answer, they look things up online and cite articles from websites. This part is our own inference: to cite something, an AI also needs a source it trusts; if an article’s author can’t be found anywhere online, the AI is less likely to rely on that article, and so has fewer chances to bring you up when it answers a customer.

What to do about it

Act on this now. Google has put it in writing, and the first step is finding out whose names are on the articles on your site. The principles fit in three lines: the person named should be real, their credentials should be checkable, and the content should contain facts only your business could write. How you choose your ingredients, or what customers ask you most: no agency or AI could guess those. AI can help with the writing, but the experience and expertise have to be your own.

Taking the fake author down, though, doesn’t make the articles genuine. Take that online shop: all twenty articles are written in Dietitian Lin’s voice from start to finish, with lines like “I often see this in clinic…” and “I usually recommend…”. Remove the name and nobody stands behind those words; put a real colleague’s name on them and they don’t have that clinical experience either. To make the articles hold up, every piece of experience and advice has to trace back to real people and records in your company, and whatever doesn’t match has to be rewritten. And if the writing has passed through several agencies, Dietitian Lin may not be the only name on your site, and the voice and back story under every name have to be checked article by article.

We can take on this kind of clean-up from the start; it’s part of our managed GEO service. Our team fixes your site’s content, brings what’s said about you elsewhere into line (for example, making your business descriptions on each platform match your website), and regularly checks how each AI engine describes you. You just review the results. The point is for AI assistants such as ChatGPT and Gemini to have the right information about you when customers ask. We can also write your new articles, and the experience and figures in them still have to come from real people and records in your company; if someone on your team wants to do the writing, they can hand us the draft, and we’ll edit it and publish it for you.

Who wrote something affects whether customers trust you; when a customer asks an AI “Which business in this trade can I rely on?”, what they want to know is who to trust. To see how AI answers that question today, pencil in time for a demo, a session with one of our consultants where they show you live. On the day, the consultant rewords that question the way people in your trade would ask it and tries it on ChatGPT (for a supplement shop, something like “Which online supplement shop can I trust?”), with your business name nowhere in the question. If you’re among the few it recommends, you’ll go through what it says about you together, checking whether the experience it credits you with is experience you really have.

Once the AI questions are done, the consultant turns to your own website and shows you its GEO audit score: geoweb.tw’s free website audit, which checks how well AI can understand your site and scores it across 12 areas.

Further reading