Google now calls AI headshots and fake bylines deception

Google’s people-first content guidance, last updated 2026-10-01, now tells site owners that inventing the person behind an article counts as deception. The new paragraph sits in the “Who (created the content)” section, directly after the existing advice to publish accurate authorship information:

“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.”

Search Engine Journal’s piece compared the live page against a copy saved June 8, which runs from the bylines advice straight into the “How” section with nothing in between. The same edit also added a section defining main content. Google changed its separate AI content guidance the same day, adding an effort and originality test to those docs.

None of the substance is new to Google’s quality raters. Section 4.5.3 of the Search Quality Rater Guidelines already lists AI-generated content with made-up author profiles, built from AI images or deceptive bios, as deception, along with profiles that falsely claim credentials such as posing as a medical professional, and tells raters to give those pages the Lowest rating. What changed is who the instruction is written for. Google’s page itself notes that search raters “have no control over how pages rank,” while the new paragraph names the company’s automated quality systems instead. It does not say how those systems would identify a fabricated profile, and no public evidence says they can do it reliably.

A genuine byline earns no ranking lift, and a fabricated one is now named as a low-quality signal. Search Engine Journal’s piece notes that Google’s Search Liaison account said in January 2024 that bylines do not help pages rank better, and nothing in the October edit walks that back. The safer read is to leave content unsigned rather than sign it with a person who does not exist, which reverses how many sites have treated Google’s E-E-A-T advice (experience, expertise, authoritativeness, trust) over the past few years.

What to do

Audit author pages and bylines for whether each name belongs to a real, verifiable person:

• Photos. Replace AI-generated or stock portraits with real photographs of the named person, or publish no photo. A stock portrait under a real staff name is still a picture of someone other than the author.
• Names. Remove pen names invented to make machine-assisted content look human-written. A house byline or no byline is defensible; a fictional expert is not.
• Credentials. Keep only claims a reader can check, and link to the registry, profile, or published work that confirms them. Medical titles are the specific example the rater guidelines give.

AI assistance in the writing itself is not the target here. Google’s objection is to the claim about who wrote the page, not to the tooling behind it.

Watch out for

The deception section of the rater guidelines is wider than authorship. It also covers fake business details, impersonation, and deceptive page design, so a site cleaning up author bios while running an invented company address has only done part of the job.

Google’s documentation updates page logged the Oct. 1 change to its AI content guidance but, as of Search Engine Journal’s publication, not this people-first content edit. There is no notification to rely on for further wording changes to that page, so check it directly.