Google: optimizing for AI search is still just SEO

Google has used its Search Central Live event in Milan to repeat a message many brands have been slow to hear: there is no separate playbook for getting cited in AI search, including Google’s AI Overviews and AI Mode. The tactics sold as “AEO” or “GEO”, such as breaking content into tiny chunks or publishing llms.txt files, do not help, Google says. What works for AI features is the same thing that works for search: genuinely useful, distinctive content.

What happened

At Search Central Live Milan on 19 June 2026, Google reiterated the guidance in its official documentation on optimizing for generative AI features, published in May 2026. Google’s position is blunt:

“There’s no requirement to break your content into tiny pieces for AI to better understand it.”

Google Search Central, generative AI optimization guidance

The same guidance says you do not need to create special machine-readable files such as llms.txt, AI-specific markup, or rewritten “AI” versions of pages, because Google Search does not use them. Google frames the whole category plainly: from its perspective, optimizing for generative AI search is optimizing for the search experience, and is therefore still SEO. What it says does matter is unchanged: unique, non-commodity content with a first-hand point of view, a crawlable and indexable technical structure, and high-quality images and video.

Why it matters

A market of tools and services has grown up around the idea that AI search needs its own optimization tricks. Google’s guidance pulls the rug from under most of them. For brands, the value is in what it removes: time and budget spent on chunking, llms.txt, and “AI rewriting” can go back into the fundamentals that actually influence visibility.

One distinction is worth keeping in view. This guidance is about how Google chooses and presents content in AI features. It does not soften the growing legal accountability for what AI then says about businesses. Getting cited and being described accurately are two different jobs.

What this means for multi-location brands

At enterprise scale the temptation is to buy a shortcut, because doing the fundamentals across thousands of pages is hard. Google’s guidance says the shortcut is not real. The work that pays off is the same work that has always paid off, done consistently across the whole estate: distinctive content on location and brand pages rather than templated boilerplate, clean and crawlable technical structure, and accurate, consistent data about every location so the systems that feed AI answers have something trustworthy to draw on.

That last point is where local brands have an advantage. The verifiable facts about each location, such as name, address, hours, and services, are exactly the kind of accurate, first-hand information AI features prefer. Treating AI visibility as an extension of SEO rather than a separate discipline, and keeping the underlying location data correct everywhere, does more for appearing in AI search results than any gimmick.

The bottom line

Google has said it plainly: there is no AI-search cheat code, just SEO done well. For multi-location brands, that means the budget belongs in content quality and accurate location data, not in tactics built to game a system that, Google says, does not work the way the tactics assume.