Google has told site owners directly that inventing an author is deception, in an edit to its page on creating helpful, reliable, people-first content. Search Engine Journal’s Matt G. Southern reported on October 2, 2026 that the version Google dated October 1 carries the new paragraph. Sites that publish under invented personas or synthetic portraits now have the warning in the document written for them.

The paragraph reads: “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.”

According to Southern, it sits in the “Who (created the content)” section, directly below the existing guidance on crediting real authors with bylines. A snapshot of the page from June 8 moves from that byline guidance to the “How” section with nothing in between. The October 1 version also adds a section defining main content, which Southern notes but does not detail.

The idea itself is not new. The Search Quality Rater Guidelines already treat invented authors as deception, in section 4.5.3. One example there is machine-written pages carrying fictional author bios and synthetic pictures. Another is an author who claims qualifications they lack, such as someone pretending to be a doctor. Raters assign the Lowest rating to pages that use deception. Roger Montti reviewed that guidance for Search Engine Journal in May 2025.

What changed is the audience. The rater guidelines speak to evaluators, and the people-first page itself says raters “have no control over how pages rank.” The new paragraph instead points at Google’s “automated quality systems.” Southern points out that the paragraph leaves open the method those systems would use to spot a fake persona. That gap matters: Google has described the offense without describing detection, and no measurement of any ranking effect has been published.

Treat this as documentation, not an announced ranking change. Google has not said that sites with invented authors will be demoted, and the source does not connect the edit to any update. Southern also points back to January 2024, when the Search Liaison account for Google stated that bylines are not a ranking boost. Read together, the message is narrow: a byline earns nothing by itself, and a false one is called a low-quality signal.

This is also a separate document from the generative AI guidance Google refreshed the same day, which we covered earlier. That page concerns reviewing AI output. This one concerns who is credited with it.

For sites running personas, the practical question is whether every credited name can be defended. An author audit should check five things:

Southern adds that the deception section of the rater guidelines reaches further than this paragraph, extending to invented business information, pretending to be someone else, and misleading page layouts, so an audit that stops at bylines leaves other exposure untouched.

One more caveat: Southern reported that Google’s documentation updates page, as of publication, logged the October 1 change to the AI content guidance but did not list this edit. Wording could still shift, so keep a dated copy of the paragraph for your records and check the page again before finalizing policy.

Teams with invented or stock-photo authors should replace them with real contributors, or publish without a persona, before treating any other optimization as the priority.

Based on Search Engine Journal reporting by Matt G. Southern, published October 2, 2026, on Google’s “Creating helpful, reliable, people-first content” page.