TLDR
An Ahrefs study led by Ryan Law found heavily machine-written pages appearing across all positions in Google's top ten results, with the average share of machine-written text rising only slightly from 27.1% at position one to 30.9% at position ten. Google appears to filter for content quality rather than AI authorship, with 82.2% of pages in the top three positions containing under 50% machine-written text. A separate indexing gap exists: pages lightly flagged for machine-written text are indexed at 49.28%, compared with 40.35% for heavily flagged pages. Law said Google penalises bad content, not AI content, though the two overlap often enough to cause real-world confusion.
KEY TAKEAWAYS
What the numbers actually show
Google does not ban AI-written pages from its top results. The average share of machine-written text rises only slightly across page one, from 27.1% at position one to 30.9% at position tenverifiedVerified Source: ahrefs.com, according to an Ahrefs study published on 27 July 2026.[1] Heavily machine-written pages appear at every rank on the first page, not just at the bottom.
Ahrefs measured machine-written text shares using its own detection tool, categorising pages from low (under 20%) to very high (over 80%), then mapped those scores against Google's ranking and indexing outcomes.[1] Ryan Law, Ahrefs Director of Content Marketing, led the study.
Quality, not origin, drives the filter
54.7% of pages in the top three Google results contained under 20% machine-written textverifiedVerified Source: ahrefs.com, and 82.2% contained under 50%.[1] That skew toward lower machine-written shares at the top is real, but it does not mean AI-heavy pages are shut out entirely.
Ryan Law said: "I don't think Google is trying to punish AI-generated content; I think it is relying on the same old hallmarks of content quality that it always has. It's just that AI-generated content is usually lower quality than human-generated content."[1]
Google's said position supports that reading. The company says it does not penalise content based on how it is produced, focusing instead on rewarding material that is genuinely helpful to people rather than content designed purely to game search rankings.[1]
The indexing gap: a different problem
Getting ranked and getting indexed are two separate hurdles, and the Ahrefs data shows a meaningful gap at the first one. Pages flagged with under 20% machine-written text were indexed by Google at a rate of 49.28%, while pages with over 80% machine-written text were indexed at only 40.35%verifiedVerified Source: ahrefs.com.[1]
That nearly nine-percentage-point gap means a meaningful share of heavily machine-written pages never enter the rankings contest at all. For publishers relying on AI to generate volume, the indexing stage is where the first losses occur, before any ranking signal applies.
Why the confusion persists
Ryan Law said: "Based on this research, I believe that Google is not against AI content; it is against bad content, but confusion arises because AI content and bad content overlap a significant amount of the time."[1]
That overlap is the practical problem. Publishers attributing ranking drops to AI authorship may be misreading a signal that is actually about thin, repetitive or low-value output, qualities that AI tools produce at scale when used without editorial judgement.
What it means for Australian businesses
For Australian businesses writing for search, the Ahrefs data reframes the decision. The question is not whether a page was written by a person or a machine; it is whether the page offers something a customer cannot get elsewhere.[1] That test applies regardless of the tool used to produce the text.
The indexing numbers add a practical note of caution for volume-first strategies. At a 40.35% indexation rate for heavily machine-written pages, close to six in ten such pages are invisible to Google searchers before the ranking question even arises.[1]
SOURCES & CITATIONS
FREQUENTLY ASKED QUESTIONS
Does Google penalise AI-written content?
Are AI-written pages indexed at the same rate as human-written pages?
What share of Google's top three results are mostly human-written?
Zara Kincaid writes about artificial intelligence and search. Her focus is what happens to businesses when the front page of the internet stops being a list of links and starts being an answer.
