iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
AI search has not made SEO fundamentals obsolete, and Google’s own documentation asks for no separate AI playbook, no special markup, and no new content format. Whether AI features are reducing the traffic your pages receive is a separate question, and the honest answer is that the evidence is mixed and depends on who measured what. In August 2025 Google said aggregate organic clicks from Google Search had stayed relatively stable, while two 2026 study preprints found fewer clicks in the settings they examined.
Everything below concerns Google Search’s AI features, including AI Overviews and AI Mode. The evidence does not establish how ChatGPT, Perplexity, or other assistants choose and cite sources, so do not carry these conclusions over to them without checking each product’s own documentation.
What Google’s documentation requires, and what it does not
Myth 1: AI search needs a separate SEO playbook
Google Search Central’s AI features documentation states: “The best practices for SEO remain relevant for AI features in Google Search (such as AI Overviews and AI Mode).” The same documentation says there are no additional appearance requirements and no special optimizations specific to these features.
The prerequisites Google points to are the ordinary ones: crawl access, index eligibility, useful page content, internal discoverability, and standard technical SEO. Meeting them does not guarantee that a page appears in an AI Overview or AI Mode. It makes the page eligible in the same way it would be for ordinary results.
Myth 2: llms.txt or special AI markup is required
Google says Search does not use llms.txt or other special AI text files to determine visibility, and it says no special schema.org markup is required for generative AI features. These two statements are separate from structured data in general. Ordinary structured data can still support rich-result eligibility for the rich-result types a page qualifies for, so keep using it for that purpose.
A site may publish an llms.txt file for other AI systems that read it. That is a decision about those systems. Google says the file has no effect on visibility or rankings in Google Search, so it should not be treated as an SEO requirement.
Myth 3: Pages must be split into chunks or rewritten in AI-specific language
Google’s guidance says there is no requirement to chunk content and no need to rewrite pages just for AI systems. It recommends writing for the audience and says there is no ideal page length. Google does not name a target length, so the right length is whatever fully answers the reader’s question. Chunking and AI-specific phrasing have no documented basis as requirements for Google’s features.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Myth 4: Every AI-generated page is automatically spam
Google’s generative AI content guidance states: “Generative AI can be particularly useful when researching a topic, and to add structure to original content.” The same guidance recommends accuracy, quality, relevance, and manual fact-checking of anything AI helped produce.
The enforcement line is value and scale, not authorship. Google says generating many pages without adding value may violate its scaled content abuse policy. A site that uses AI to draft pages it fact-checks and improves is in a different position from one that mass-produces pages with little added value. The policy looks at what the output offers readers, not whether a model helped write it.
What the traffic evidence shows
Myth 5: AI Overviews have a single, proven traffic effect
On August 6, 2025, Google stated that aggregate organic click volume from Google Search to websites was relatively stable year over year, and that average click quality had increased. This is Google’s own characterization of its platform. It is not an independently replicated result, and the passage did not provide a specific percentage or the methodology behind the claim, so no figure should be attached to it.
Rank #3
Two points follow. An aggregate can hold steady while individual queries, sites, or categories lose clicks, so “stable in aggregate” does not mean “unaffected.” And “click quality increased” describes Google’s view of the clicks it observes. It does not establish that visitors arriving from AI features are higher quality for every site.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsMyth 6: User-level studies show no click or browsing effect
Two 2026 study preprints, identified by arXiv number, cut against the idea that AI Overviews leave clicks unchanged. Both are bounded by their own samples and designs.
The observational study, arXiv:2608.04831, analyzed one month of Google searches from a representative panel of 900 U.S. adults. Its reported results were:
Rank #4
- 18% of searches generated an AI Overview. This is the panel’s observed share for that month, not a general prevalence estimate for all users or dates.
- About 1% of AI Overview visits included a click to a cited source. This measures cited-source clicks only, not every kind of click from the page.
- 26% of pages with an AI Overview ended the user’s browsing session, compared with 16% of pages without one. This is an association. It does not prove the overview alone caused every session ending.
The preregistered field experiment, arXiv:2608.18352, involved 1,100 participants and reports that removing AI Overviews and AI Mode increased clicks to publishers. Because it compares treatment and control conditions, it supports a causal reading within its own setup. It does not give the size of the effect for your site, your queries, or your market.
“No effect” is therefore untenable for the settings these studies examined. What they do not establish is a universal percentage loss for every publisher, query, or country.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Comparing the three kinds of traffic evidence
These sources answer different questions, so they should not be merged into one “AI search traffic loss” number. Before you quote any traffic figure, check who produced it, what it measured, its scope, and whether it supports a causal reading.
| Source | Who produced it | What it measured | Scope and period | Causal reading | Reported result |
|---|---|---|---|---|---|
| Google statement, August 6, 2025 | Google, as a platform statement | Aggregate organic click volume and average click quality | Google Search overall; sample and methodology not stated in the passage | Not a causal test; platform characterization | Volume relatively stable year over year; click quality increased; no percentage given |
| arXiv:2608.04831 (2026 preprint) | Study authors, observational design | Searches with AI Overviews, cited-source clicks, session endings | One month; representative panel of 900 U.S. adults | Association only | 18% of searches generated an AI Overview; about 1% of AI Overview visits included a cited-source click; 26% of pages with an Overview ended the session versus 16% without |
| arXiv:2608.18352 (2026 preprint) | Study authors, preregistered field experiment | Publisher clicks when AI Overviews and AI Mode were removed | 1,100 participants; period and markets not stated | Controlled comparison within its design | Removal increased publisher clicks; magnitude not stated |
Measuring AI feature traffic
Myth 7: Third-party tools can reveal Google’s internal AI ranking signals
Google Search Central states that no third-party tool has access to its internal ranking or AI systems. A tool’s visibility score or citation estimate is a model built from outside observation, not a readout of Google’s signals. Such metrics can still help organize monitoring work, provided you label them as estimates and check the tool’s advice against Google’s documentation before acting on it.
Reading AI feature traffic in Search Console
Google’s AI features documentation says appearances in AI Overviews and AI Mode are counted within the overall Web search type in Search Console’s performance reporting. It does not describe a separate search type for them. A drop in Web clicks may therefore come from AI features, from ranking changes, or from both, and the report alone cannot separate the two.
Pair the Search Console figures with conversions, engagement, and time on site from your analytics. A click count is one input to whether a page is doing its job, not the whole outcome.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
