Fake Author Profiles: Google Now Calls Them Deception, and What to Fix First
Google’s guidance on people-first content now says to “avoid using deceptive authorship information,” naming AI-generated headshots, made-up names and false credentials as examples. The page is dated 5 October 2026, and the wording was not there on 24 September.
Bylines still do not help a page rank. A fabricated byline is now named as a signal of a low-quality page. Here is what the paragraph says, where it came from, what it means for AI search, and how to show a real author so that both readers and machines can find them.
What did Google add about deceptive authorship?
Google’s page Creating helpful, reliable, people-first content has long asked three questions under the heading “Who (created the content)”: is it self-evident who authored the content, do pages carry a byline where one is expected, and do bylines lead to more information about the author. It then says Google strongly encourages “accurate authorship information, such as bylines to content where readers might expect it.”
The new paragraph follows directly: “However, avoid using deceptive authorship information. Fabricating creator profiles (such as by using AI-generated headshots, made-up names, or false credentials to make content appear as if it was written by human experts) is a form of deception. Any form of deception makes a page untrustworthy to both users and our automated quality systems, and is a signal of a low-quality page.”
Three things stand out. The examples are specific: AI headshots, invented names, false credentials. The test is intent: the profile is fabricated “to make content appear as if it was written by human experts.” And the consequence is stated in Google’s own ranking language: untrustworthy to its automated quality systems, and a signal of a low-quality page.
When did the paragraph appear, and did Google announce it?
The page carries the footer “Last updated 2026-10-05 UTC.” A Wayback Machine capture from 24 September 2026 shows the same page dated 10 December 2025, with no mention of deceptive authorship. So the wording was added between 24 September and 5 October 2026.
Google did not announce it. Its documentation change log lists two October entries, the 1 October update to the generative AI content guide and the 8 October UGC Fresh Data Program page, and neither mentions this page. Search Engine Journal reported the paragraph on 2 October, when the page was dated 1 October; Search Engine Roundtable covered it on 6 October. We wrote up the 1 October change in what Google’s AI content guidance says now. The same edit also added a section defining “main content” and the four attributes raters evaluate it on: effort, originality, talent or skill, and accuracy.
The ideas are not new to Google. Search Engine Journal notes the Search Quality Rater Guidelines already told raters to treat made-up author profiles, AI-generated author images and false claims of expertise as deception, and to give such pages the Lowest rating. What changed is the audience: the warning now sits in guidance written for site owners, not only in a document for raters.
Does a byline help you rank?
No, and Google has said so twice. In January 2024 Google’s Search Liaison account posted, in a thread Search Engine Journal quoted at the time: “Author bylines aren’t something you do for Google, and they don’t help you rank better. They’re something you do for your readers — and publications doing them may exhibit the type of other characteristics our ranking systems find align with useful content.”
Earlier, in August 2021, when Google began supporting the author URL property in Article structured data, Barry Schwartz asked John Mueller whether people would fake author details. Mueller’s reply, recorded by Search Engine Roundtable: “We use a lot of signals in Search, so I’m not too worried about spam yet.”
Read together with the new paragraph, the position is consistent. A real byline earns nothing by itself. A fake one is now named as a low-quality signal. The upside is for readers; the downside is for rankings.
Why do fake authors matter more in AI search?
Because AI answers are built from attributed passages. When ChatGPT, Perplexity or Google’s AI Mode cite a page, they lift a chunk and name its source. Our generative engine optimization work is built on making each page quotable: a named person, a verified date, sources the engine can follow. An invented expert breaks the chain at the first link. If the person does not exist, the credential cannot be checked, the profile page leads nowhere, and the page fails Google’s own “Who” test.
Google’s page is explicit that trust is the most important part of E-E-A-T, and that its systems give more weight to strong E-E-A-T on topics that affect health, money or safety. Real estate, legal services, finance and medical practices, the businesses most tempted to add a stock “expert” to a blog, are exactly the topics where the fabricated profile costs the most.
There is also a practical cost nobody writes about. An AI-generated headshot of a person who does not exist cannot sit on a LinkedIn profile, a state licence lookup or a Google Business Profile. The moment a prospect searches the name, the gap shows. The engines are catching up to what customers already do.
What counts as fabricated, and what is fine?
Google’s text gives three examples. Here is what each looks like on a typical business site, and the fix.
| Google’s example | How it shows up on business sites | Fix |
|---|---|---|
| AI-generated headshots | A “Senior Editor” or “Market Analyst” on the blog whose face was generated, with no profile anywhere else | Use a real photo of a real person, or no photo and a company byline |
| Made-up names | A content vendor’s invented author rotated across many client sites | Put the owner, a licensed team member or the company name on the byline |
| False credentials | “Licensed broker”, “CPA” or “MD” added to a bio that cannot be verified | State only credentials that a public register or profile can confirm |
What the page does not say also matters. It does not ban AI-assisted writing; Google’s 2023 statement that it rewards quality “rather than how it is produced” still stands. It does not require an author on every page; it says bylines belong “where readers might expect it.” And it does not address pen names or company bylines. A post signed by the business, with a real About page behind it, is honest. A post signed by a person who was never born is the problem.
How many Miami real estate sites show who is behind them?
We looked. On 12 October 2026 we fetched the homepages of the 70 Miami residential agent, team and brokerage sites from our September AI-readiness study; 65 answered. We did not judge whether the people named are real. We checked whether a visitor, or a machine, can find who is behind the site from the front door.
- 46 of 65 link to an About, Team or Agents page from the homepage.
- 12 of 65 carry Person structured data on the homepage.
- 8 of 65 carry any author property in their structured data.
Most of these sites are run by real, licensed people, so the gap is not deception. It is the opposite problem: a real author that machines cannot see. Google’s “Who” questions ask whether bylines lead to further information about the author. For most of this sample the answer, in the markup, is silence. That is the easier half of the fix, and it is the half our real estate marketing clients start with.
How do you mark up a real author correctly?
Google’s Article structured data documentation includes a short list of author markup best practices. Follow it in this order:
- Use the right type. Set the author’s
@typeto Person for a human and Organization for the company. Google says it strongly recommends including the type and url fields. - Name only the name. The
author.namefield holds the person’s name and nothing else: no job title, no publisher, no “Team” suffix. Titles go injobTitle. - Link to a page that identifies the author.
author.urlshould point at a profile that uniquely identifies the person, for example an internal bio page or a social profile. If it is an internal page, Google recommends marking that page up with profile page structured data. - List every author, one per field. Everyone shown as an author on the page must be in the markup, each in their own author entry, never merged into one name string.
- Make the visible byline match. The name on the page, the name in the schema and the person on the About page must be the same person. That is what we do here: every post is signed by Josh Stein, links to the agency page, and carries the same name in the BlogPosting schema.
One caution from Google’s own page: raters “have no control over how pages rank” and rater data is not used directly in ranking. The paragraph refers to Google’s automated systems, and does not say how they identify a fabricated profile. Treat the markup as a way to be understood, not as a ranking lever.
What should a business owner do this week?
Four steps, none of which needs a developer for more than an hour.
- Open your own blog and read the bylines. If you do not recognise a name, ask your content vendor who it is. If the answer is a stock photo and a pen name, replace it with yourself or the company.
- Check every credential you publish. Licence numbers, designations and degrees should match a public register. A sales associate is not a broker; a consultant is not an attorney.
- Give each real author a page. A bio, a real photo, the licence, and links to the person’s own profiles. Point the byline and the schema at it.
- Check what machines see. Run a page through the free AI Visibility Checker, which reports whether a named author, a date and sources are readable, and see how we run agentic marketing with a human name on every output.
If you publish with AI, the earlier guidance still applies: a person fact-checks every page before it goes live. The new paragraph adds one line to that process. The person on the byline has to be real too.
Want your business found on Google and in ChatGPT? Ciao Bella engagements start at $5,000 a month, no setup fee. See how we work.
Verified 12 October 2026 against Google Search Central’s people-first content page, documentation change log and Article structured data guide, the 24 September 2026 Wayback capture, and our own homepage check of 65 Miami real estate sites
Building on the web since 1999 · South Florida and Mexico City
Frequently asked questions
Does Google penalize AI-generated author photos?
Google’s page does not describe a penalty. It says fabricating creator profiles, including with AI-generated headshots, is a form of deception that makes a page untrustworthy to its automated quality systems and is a signal of a low-quality page.
Is an author byline a Google ranking factor?
No. Google’s Search Liaison said in January 2024 that bylines do not help a page rank better and are something you do for readers. The new paragraph adds that a deceptive byline is a low-quality signal.
Did Google announce the deceptive-authorship change?
No. The page footer reads last updated 5 October 2026, but Google’s documentation change log lists only the 1 October generative AI guidance update and the 8 October UGC Fresh Data Program page. Trade press reported the paragraph on 2 and 6 October.
Can I publish blog posts under my company name instead of a person?
Google’s page asks for accurate authorship where readers expect it and says the author in markup can be a Person or an Organization. A company byline with a real About page is accurate. A named person who does not exist is what the page calls deception.
Do I need author structured data on every page?
Google recommends author markup for articles, with the type and url fields, the name field holding only the name, and every visible author listed separately. Service and product pages usually need an Organization, not a Person.
Sources
- Google Search Central — Creating helpful, reliable, people-first content Last updated 5 October 2026. Source for the deceptive authorship paragraph, the Who/How/Why questions, the main-content definition and the rater note.
- Wayback Machine — the same page captured 24 September 2026 Page dated 10 December 2025; no deceptive authorship paragraph.
- Google Search Central — Latest documentation updates October 2026 entries: 1 October (generative AI guidance) and 8 October (UGC Fresh Data Program). No entry for this change.
- Google Search Central — Article structured data, author markup best practices Source for the Person/Organization type, name-only rule, author.url and one-author-per-field guidance.
- Google Search Central Blog — Google Search and AI-generated content (February 2023) Google’s statement that quality matters more than how content is produced.
- Search Engine Journal — Google: Author Bylines Not A Ranking Factor (10 January 2024) Quotes the Search Liaison post on X that bylines do not help pages rank.
- Search Engine Journal — Google Adds Fake Author Warning To Helpful Content Guidance (2 October 2026) First report of the paragraph; notes the rater guidelines already described fake profiles as deception.
- Search Engine Roundtable — Google Warns Against Deceptive Authorship Information (6 October 2026) Report of the added paragraph with a screenshot of the section.
- Search Engine Roundtable — Google Not Yet Worried About Faking Author Details (11 August 2021) John Mueller’s on-the-record reply about author spam.