What Google actually says about ranking in AI answers

Why this matters now

Your buyers have started asking ChatGPT, Perplexity and Google's own AI Overviews who to trust before they visit your site. A cottage industry has sprung up around this shift, selling "GEO" and "AEO" as a distinct, almost mystical discipline with its own secret rules. Some of that is real. A lot of it is not.

Google publishes a plain-language guide on exactly this topic, aimed at site owners: Google's Guide to Optimizing for Generative AI Features, alongside a companion reference, AI Features and Your Website. Almost nobody reads either one. This is what they say, section by section, with no padding.

The core principle: it's still SEO

Google is direct about this. In its own words: "From Google Search's perspective, optimizing for generative AI search is optimizing for the search experience, and thus still SEO." AEO and GEO are terms you'll encounter, but Google frames them as still falling under SEO, not a separate discipline. AI Overviews and AI Mode are not a separate ranking system with a separate rulebook. They pull from the same index, the same crawl, the same quality systems as regular Search results.

That single sentence undercuts most of what's sold as "GEO strategy." There is no separate AI-search index to game.

How AI Overviews and AI Mode actually pull content

Two mechanisms do the work, per Google's own definitions:

• Retrieval-augmented generation (RAG), also called grounding: "a technique used to improve the quality, accuracy, and freshness of AI responses by relying on our core Search ranking systems to retrieve relevant, up-to-date web pages." The system pulls specifics from those pages and generates a response with links back to the source.
• Query fan-out: "a set of concurrent, related queries generated by the model to request more information and fetch additional relevant search results to address the user's query." This is why depth and breadth of genuine coverage on a topic matters more than a single perfectly-optimized page.

What Google says actually helps

The guide is explicit that there are "no additional requirements to appear in AI Overviews or AI Mode." Standard fundamentals apply, none of them exotic:

• Non-commodity content. A real point of view or first-hand experience, not a restatement of what already ranks. Google's own example contrasts a generic "7 Tips for First-Time Homebuyers" post with a non-commodity piece like "Why We Waived the Inspection & Saved Money: A Look Inside the Sewer Line" — a genuine expert or experienced take beats a rehash anyone could write.
• Clean technical structure. Indexed, snippet-eligible, crawlable, semantic HTML, unblocked JavaScript, minimal duplication, and a page experience that doesn't get in the way.
• Local and ecommerce feeds kept current, where relevant (Merchant Center, Business Profile).
• Monitoring through the Generative AI performance report in Search Console, so you can see what's actually happening instead of guessing.

The GEO myths Google explicitly debunks

This is the part that should worry anyone paying for "AI optimization" packages. Google names these directly, in the same guide, and says they don't help:

• llms.txt and other "AI text files." "You don't need to create new machine readable files, AI text files, markup, or Markdown to appear in Google Search." Creating one neither harms nor helps visibility.
• Chunking content into tiny fragments so AI can supposedly parse it better. "There's no requirement to break your content into tiny pieces for AI to better understand it."
• Writing in a special "AI-friendly" style. "You don't need to write in a specific way just for generative AI search." The systems understand synonyms and general meaning. There's no secret phrasing that ranks better.
• Chasing inauthentic "mentions" across the web to manufacture AI citations. Core ranking systems focus on quality; other systems are built specifically to catch this kind of manipulation.
• Special schema markup "for AI." "Structured data isn't required for generative AI search, and there's no special schema.org markup you need to add." It's still worth using for standard rich results, just not for this reason.

The controls you already have

Less discussed, and genuinely useful: Google gives site owners real levers over how AI features use a page, laid out in its robots meta tag documentation.

• nosnippet, data-nosnippet, max-snippet — limit whether, or how much, a page can be quoted in a snippet or AI-generated answer. Google's spec is explicit that these directives apply "to all forms of search results (such as Google web search, Google Images, Discover, Assistant, AI Overviews, AI Mode) and will also limit how much of the content may be used as a direct input for AI Overviews and AI Mode."
• noindex — keep a page out of Search, and therefore out of AI features, entirely.
• Google-Extended — a separate robots.txt directive that limits whether a page is used to train or ground Gemini models, without affecting your normal Search ranking: "Google-Extended does not impact a site's inclusion in Google Search nor is it used as a ranking signal in Google Search."

Part of a proper audit is checking whether any of these are set by accident, quietly suppressing a page you actually want AI systems to cite. I go through the eligibility requirements and each of these controls in more depth in a companion guide: how to appear in AI Overviews and AI Mode.

What this means for your strategy

Getting recommended by ChatGPT, Perplexity, or Google's own AI Overviews is mostly the fundamentals, done properly, on content that actually says something. The teams selling GEO as a hidden formula, a special file, or a secret writing style are, by Google's own account, selling nothing.

This lines up with how the AI-search shift played out on one real site I worked on: rankings and impressions kept climbing while click-through rate quietly fell, the exact signature of AI Overviews absorbing a query before it reached the site. The fix wasn't a GEO trick. It was restructuring genuinely good content so an AI system could lift the answer cleanly and trust the source enough to cite it. The full case study is here, numbers included.

Sources: Google's Guide to Optimizing for Generative AI Features and AI Features and Your Website, Google Search Central. Accessed July 22, 2026.