Google Is Cracking Down on AI Content. Here’s What Works
Summary
Google revised its "Creating helpful, reliable, people-first content" guidance on October 1, 2026. The page now tells Search Quality raters to judge a page's main content on four criteria: effort, originality, talent or skill, and accuracy. Google gives using generative AI to produce large amounts of text without oversight as an example of low effort, and it calls fabricated author profiles deceptive.
In a Surfer Academy video, Matt Kenyon explains the change. He describes a "rank and tank" cycle hitting AI-scaled sites across Google, AI Overviews and ChatGPT. He argues that ChatGPT gives vendor listicles less weight than independent reviews, and he lays out 7 principles for content that holds up, using examples such as Outdoor Gear Lab and State Farm. He also cites Surfer research across about 5 million AI citations, which found that domain authority had almost no correlation with whether a page got cited.
Demonstration by Surfer Academy, embedded from YouTube. AIOTruth did not reproduce the test.
Why it matters
Brands that publish AI-generated pages in bulk to win visibility now face a written quality standard that rewards effort, originality, skill and accuracy, and that names unsupervised AI text and fake authors as problems. For anyone trying to be cited or recommended by AI systems, the practical shift is toward fewer, deeper pages written by real, named authors and backed by firsthand work. If self-promotional listicles lose weight in ChatGPT, a brand earns recommendations through independent reviews and original material, not by ranking itself first on its own site.
Source
- developers.google.com/search/docs/fundamentals/creating-helpful-content the item itself, published by Surfer Academy, created by Surfer Academy
- lilyraynyc.substack.com/p/prediction-the-next-massive-google Reddit or YouTube discussion
- www.hobo-web.co.uk/seo-blog/ Reputable reporting
- youtube.com/watch?v=hi9ahSsMfj8where AIOTruth found it
What was checked
- canonical URL taken from the page itself
- read the item page (200)
- https://services.google.com/fh/files/misc/hsw-sqrg.pdf: does not mention the item, tier 8 (Named expert analysis, or a first-party claim about itself)
- https://developers.google.com/search/docs/fundamentals/creating-helpful-content: supports the item, tier 1 (Official platform documentation)
- https://developers.google.com/search/docs/fundamentals/creating-helpful-content?hl=ar: does not mention the item, tier 1 (Official platform documentation)
- https://static.googleusercontent.com/media/guidelines.raterhub.com/en//searchqualityevaluatorguidelines.pdf: does not mention the item, tier 9 (Reputable reporting)
- https://lilyraynyc.substack.com/p/prediction-the-next-massive-google: supports the item, tier 10 (Reddit or YouTube discussion)
- https://www.hobo-web.co.uk/seo-blog/: supports the item, tier 9 (Reputable reporting)
Limitations
- The demonstration was not independently reproduced by AIOTruth.
AIOTruth judgment
The item earns its Confirmed status because the central change is on Google's own Search Central page, a first-party source, which supports the evidence score of 10. Lily Ray's analysis and the Hobo SEO Blog add context on the documentation update. Usefulness scores 9 because the guidance translates directly into editorial decisions about authorship, effort and AI oversight. Originality scores 9 because the video pairs the policy change with observed patterns in AI answers and Surfer's own citation research. Relevance scores 7 because the core document governs Google Search quality, and its link to AI citation rests mainly on the video's commentary. The Surfer citation findings and the live walkthrough come from the publisher, and AIOTruth did not reproduce the demonstration.
How this score was calculated
| Dimension | Weight | Score | What produced it |
|---|---|---|---|
| AIO relevance | 30% | 7 | ai visibility, ai overviews, ai citation; 4 scope question(s) matched |
| Usefulness | 30% | 9 | 2 artifact(s), 5 actionable marker(s), 1 measured figure(s) |
| Evidence | 25% | 10 | Official platform documentation; 3 verified source(s) across 3 domain(s) |
| Originality | 15% | 9 | carries original testing or data; closest archive match 0.1 |
aioRelevance x 0.30 + usefulness x 0.30 + evidence x 0.25 + originality x 0.15. The rubric is published in full on the Editorial Method page. Scoring is deterministic: the same item scores the same on every run.
Review the documentation Open the source
Found something wrong here? AIOTruth corrects material errors openly. Have something we should review? Submit a find.