Fake Author Bios: Google's New Warning and How to Fix Yours

Google's "Creating helpful, reliable, people-first content" page now tells site owners to avoid deceptive authorship, naming AI-generated headshots, made-up names and false credentials, and states that "any form of deception makes a page untrustworthy to both users and our automated quality systems." The page shows a last-updated date of 5 October 2026. If your site has authors who do not exist or credentials you cannot prove, fix them this week. (As of 9 October 2026.)

Key Takeaways

• Google's page names fabricated authorship examples directly: made-up names, AI-generated headshots, false credentials.
• The page also asks whether pages carry a byline where one might be expected, and whether bylines lead to further information about the author.
• No source I reviewed shows a new penalty or a new detection method; the wording is a clear quality signal, not an announced enforcement switch.
• The fix is not "delete all bylines". It is to use real, verifiable people, or an honest organisation byline, and to link bylines to real author pages.
• Do it alongside the post-spam-update review, since the September 2026 spam update finished on 8 October and thin, anonymous content is the usual overlap.

What Google changed, and what is only reported

The primary source

I opened Google's people-first content page directly. It currently says, among other things:

• Pages should show who created them. It asks, "Do pages carry a byline, where one might be expected?" and whether bylines "lead to further information about the author or authors involved".
• Google "strongly encourage[s] adding accurate authorship information" where readers might expect it.
• It warns you to "avoid using deceptive authorship information", with AI-generated headshots, made-up names and false credentials given as examples.
• It defines main content as "any part of the web page that directly helps the page achieve its purpose" and lists four quality attributes: effort, originality, talent or skill, and accuracy.

The secondary reports

The edit was first flagged publicly by Marie Haynes in a video dated 2 October 2026. PPC Land summarised it under the headline Google says fake author profiles signal a low-quality page. Two things from that coverage deserve care:

• PPC Land notes the wording was initially secondhand, relying on Haynes's reading. I have since confirmed the current Google page contains the deceptive authorship language, but I cannot tell you exactly what the page said before the edit.
• Haynes's suggestions that the edit may precede a core update, or that Google may rely less on user signals, are her speculation. Google has not confirmed either.

So the honest summary is this: Google has put in writing that fabricated author information is deception and untrustworthy. It has not told us about a new penalty, a new classifier or a timeline.

Why this matters now

Context: the spam update and manual actions

The edit landed while the September 2026 spam update was still rolling out; it finished on 8 October per Google's Search Status Dashboard. I have a separate post on what happened and what comes next. Some SEOs report more manual actions; see my evidence review. Fake bios are a classic accompaniment of scaled content: a site publishes hundreds of posts under invented "staff writers" to look human.

Google's spam policies do not list "fake author" as its own policy name, so I would not call a fake bio a spam violation by itself. But it can sit alongside scaled content abuse, and it undermines the trust signals that quality systems look for.

What is not a problem

Let me be careful, because I have already seen some people over-correct.

• An honest organisation byline ("Editorial team", "Company name") is not deceptive if the content really is an organisation's work. Google says to add authorship where readers would expect it; it does not say every page needs a personal name.
• AI assistance is not the issue. Per the reporting, effort is judged by the human work behind content; the deception is presenting invented people as human experts.
• Pen names are a grey area. A pseudonym used consistently by a real writer, with no false credentials, is different from an invented doctor. If your pen name implies expertise, make sure there is a real person with that expertise behind it.

The audit: five checks on every author

1. Does this person exist?

For every named author, can you name the real human, and do they know they are listed? If the answer is "it was a persona we created", you have your first fix.

2. Is the headshot a real photo of that person?

AI-generated headshots are named in Google's page. If the photo came from a generator or a stock library presented as a staff member, replace it with a real photo or remove the photo entirely. A missing photo is better than a false one.

3. Can every credential be verified?

List each claimed degree, licence, job title or "years of experience". Remove any you cannot prove. For health, finance or legal topics, the stakes are higher; the reported guidance is stricter on accuracy for content that affects health, finances or well-being.

4. Does the byline lead somewhere?

Google asks whether bylines lead to more information about the author. Create an author page per real writer: who they are, what they cover, links to their real profiles, and a list of their articles. Make the byline link to it.

5. Does the author fit the content?

A financial adviser byline on a recipe post, or a "board-certified" label on a post nobody medical reviewed, is a mismatch. Keep reviewers and writers distinct and honest: "Written by X, reviewed by Y".

A decision table for what to do with each author

How to fix it without breaking things

Step 1: Export your authors

Pull every distinct author name from your CMS with post counts. In most CMSs this is a single query or an export. Mark each one against the five checks.

Step 2: Decide the replacement byline

For posts by invented personas, you have three honest options: attribute them to the real writer, attribute them to the organisation, or have a real subject expert review and put their name as reviewer. Do not just swap one invented name for another.

Step 3: Fix structured data

If your pages use Article schema with an author property, make sure it matches the visible byline. Schema.org's Person type lets you describe a real author and link to their profiles with sameAs. Do not add Person markup for people who do not exist; markup that contradicts reality adds risk, not value.

Step 4: Update the posts, then log it

Change the byline, the schema and the author page together. Record dates in a change log. If your audit sits alongside a post-update review, use the keep/improve/merge/remove method in my content audit guide.

Step 5: Do not stop at bios

Google's attributes are about the main content: effort, originality, talent or skill, and accuracy. A true bio on a thin page does not rescue the page. Check that the page itself shows first-hand experience, original information, clear writing and accurate claims.

For agencies and publishers

If you run content for clients, put authorship in the contract. Decide who gets credited, who verifies credentials and who approves AI assistance. Ban invented personas outright. If a client already has them, the cleanup is a client conversation before it is an SEO task.

Disclosure

I am the owner of this site. My own author information is real, and my own background is stated on my About page. Where I make claims about experience I stick to what I can show.

Sources and limits

Also opened for this post: Google's Manual actions report help page, Search Engine Roundtable's report on rising manual actions and PPC Land's spam update completion report.

Primary: Google's people-first content page (last updated 5 October 2026 per the page) and spam policies. Secondary: PPC Land's report of Marie Haynes's video. I could not compare the page against its prior wording, and I have not seen evidence of a classifier that detects fake headshots.

FAQ

Does Google penalise fake author bios?

Google's page says deceptive authorship makes a page untrustworthy to users and automated quality systems. I have not seen an announced penalty specific to fake bios, so treat it as a quality and trust risk.

What counts as deceptive authorship?

Google names made-up names, AI-generated headshots and false credentials. The common thread is presenting invented people or unproven expertise as real.

Do I need a personal byline on every page?

Google encourages accurate authorship information where readers would expect it. For some content, an honest organisation byline is appropriate. The key is that it is true.

Is AI-assisted content a problem?

Not by itself, according to the reporting on the page's effort attribute. The risk is mass AI text with no human oversight, and hiding that behind invented experts.

Can I use a pen name?

A pen name used by a real writer without false credentials is a grey area but generally different from an invented expert. Avoid pen names that imply qualifications nobody holds.

Should I add Person schema for authors?

Yes, for real authors, and keep it consistent with the visible byline. Schema.org's Person type supports name, URL and sameAs links. Do not add markup for personas that do not exist.

Is this tied to a core update?

Marie Haynes suggested the edit may precede one, but that is her speculation and Google has not confirmed it. Plan around quality, not guesses.

How quickly should I fix this?

This week. It is a cheap, low-risk change compared with a content rebuild, and it removes a trust problem while you review the rest of the site.

Does this apply outside the US?

The page is general Google Search guidance, not limited to a region. I found no country-specific version of the warning.

Talk through your site's trust signals

If you want a practical review of your author pages, bylines and schema, see my work at younusfardeen.in and reach me through the contact form. I bring 4+ years of marketing experience in organic growth and AEO, and I will give you an honest list of what to fix first.