Google Confirms: AI Search Optimisation Is Still SEO
Google has published its most direct guidance to date on how businesses should approach optimisation for AI-powered search features, and the headline finding is one that will cut through a lot of noise: there is no separate discipline required.
The new documentation, published on Google’s developer site in May 2026, covers AI Overviews, AI Mode and the broader direction of generative AI in Search. It builds on an earlier 2025 guide that explained how AI features work and how content inclusion is controlled. This new version focuses squarely on what businesses should and should not do to appear in AI-generated results.
The most significant statement in the guide is straightforward. Google defines both AEO (answer engine optimisation) and GEO (generative engine optimisation) as terms the industry has adopted, then states that from its perspective, optimising for generative AI search is optimising for the search experience itself. In other words: it is still SEO.
This is not entirely surprising to those who follow Google closely. Senior figures at Google have said similar things at industry conferences over the past year. What has changed is that it is now in writing, in official documentation. Businesses and agencies can point to it directly.
What Google Says You Do Not Need to Do
The guide includes a mythbusting section that is worth reading carefully, because it explicitly names several tactics that a growing number of AI search services have been recommending.
llms.txt files and AI-specific markup. Google says there is no need to create machine-readable files, AI text files, or Markdown versions of your content in order to appear in generative AI results. Google can discover and index a wide range of file types, but that does not mean those files carry any special weight.
Chunking content. Some guides have suggested breaking content into small, discrete sections to make it easier for AI systems to process. Google says this is unnecessary. Its systems can understand the nuance of a page covering multiple topics and surface the relevant portion for any given query. Google engineers have reportedly recommended against chunking specifically.
Rewriting for AI. There is no requirement to restructure your writing or chase every long-tail keyword variation to accommodate AI systems. Google’s systems understand synonyms and general meaning. Write clearly for your audience.
Seeking artificial mentions. AI features can surface what is being said about a business across blogs, forums and video platforms. The temptation to manufacture that coverage is understandable, but Google says the approach is less effective than it appears, because its quality and spam systems operate across the same signals.
Special structured data or schema markup. Structured data remains useful for rich results, and Google recommends continuing to use it as part of standard SEO practice. But there is no new or special schema designed for generative AI features. Businesses chasing AI-specific markup are solving a problem that does not exist.
What Google Says Does Matter
The optimisation advice in the guide follows familiar territory, though with some useful framing for the AI context.
The clearest emphasis is on non-commodity content. Google draws a direct contrast between generic, widely-available information and content that offers genuine insight from real experience. The example it uses is illustrative: a listicle of homebuying tips is commodity content; a first-person account of a specific decision and its outcome is not. The distinction is whether a piece of content provides something a reader could not have found in a dozen other places.
On the technical side, the fundamentals remain the same. Pages need to be indexed and eligible for snippets. That means following standard crawling and indexing best practices, using semantic HTML where appropriate, meeting JavaScript SEO requirements, delivering a good page experience and managing duplicate content sensibly.
For local businesses and ecommerce, the guide recommends Merchant Centre feeds and Google Business Profiles as the routes to visibility in AI-generated responses. It also references Business Agent, a feature that allows customers to have conversational interactions with brands directly through Google Search.
A Note on AI Agents
The guide introduces early guidance on what Google calls agentic experiences: AI systems that can take actions on behalf of users, such as comparing products or making bookings. Google notes that browser-based agents may navigate websites by analysing screenshots and interpreting page structure, which raises some practical questions about how businesses should approach accessibility and site architecture.
The guidance here is framed as forward-looking rather than urgent. Google describes agent optimisation as something to explore if it is relevant to your business and you have capacity. It also references the Universal Commerce Protocol, an emerging standard co-developed with Shopify and endorsed by more than 20 companies, which is intended to allow Search agents to take more actions on behalf of users over time.
What This Means for UK Businesses
For UK SMEs that have been wondering whether they need to invest in a separate AI search strategy on top of whatever they already do for organic visibility, this guidance simplifies the decision considerably, for Google at least.
The core signal is clear: businesses that invest in genuine, experience-led content and maintain solid technical foundations are already building in the right direction. The tactics being sold as AI-specific requirements by some providers, including llms.txt files, content chunking and AI-specific schema, are not supported by Google’s own documentation.
That does not mean all AI platforms behave identically. ChatGPT, Perplexity and other AI tools operate with different architectures and may weight signals differently. The Google guidance settles the question for Google’s own AI features. For visibility across the broader AI search ecosystem, the approach remains more nuanced.
What is consistent across all platforms is the underlying logic: authoritative, specific, experience-led content from sources with clear expertise performs well. That has always been the direction of travel. Google’s new guide confirms that generative AI has not changed it.
AI Search Ltd helps UK businesses become visible and recommended across AI search platforms including Google AI Overviews, ChatGPT, Perplexity, Bing Copilot and Gemini. To understand where your business currently stands in AI search results, get in touch.
