CLEARWATER, Fla. — Aug. 31, 2026 — Search Engine Journal reported Pew Research findings that about one in ten live pages show signs of AI authorship, including pages carrying a post-ChatGPT date.
“1 in 10 webpages shows signs of AI authorship, Pew reports.”
— Search Engine Journal
The number will move. The direction will not surprise anyone who publishes at volume. What matters for identity is the side effect: more of the web now sounds finished without being checked.
Fluency used to be a tell
A clean About page once implied a person sat behind it. That tell is gone. Anyone can generate a complete-looking company in an afternoon. Answer engines know this even when they cannot prove it on a given URL.
Pew measured authorship signals, not honesty. A human can write a thin page. A model can write a thick one. The research is useful because it puts a floor under a feeling editors already had: prose is no longer scarce.
Why identity has to be dated
If copy is cheap, the facts that are expensive to fake have to carry more weight. A public record that says when the website was checked, when the phone was observed, and which filing number was supplied is harder to mint at scale than another generated paragraph.
That is not a claim that dated files make Pew’s number fall. It is a claim about what a reader — human or machine — should trust after the number rose.
What this brief does not say
AI Verified did not run the Pew study and does not oppose machine-assisted writing. Plenty of honest pages will be drafted with tools. The newsroom line is narrower: do not treat fluent copy as evidence that a firm exists.
Read the source: SEJ on Pew’s AI-authorship findings.