AI SEO: getting cited by AI, or using AI to do SEO

AI SEO is used for two different jobs. One is getting your business found and cited in AI answers, such as Google's AI Overviews and ChatGPT. The other is using AI tools to do SEO work faster. The first changes what you aim for, the second only changes how you work.

By AI Search Ranking. Published 21 October 2026. Last updated 21 October 2026. In AI SEO.

On this page

The short answer

AI SEO is search engine optimisation for a world where AI writes the first answer. Most of the time, it means making your business easy for AI search to find, trust and cite. That covers Google's AI Overviews and AI Mode, plus assistants such as ChatGPT, Gemini, Perplexity and Claude.

The same phrase is also used for something else: using AI tools to do SEO tasks, such as keyword research or first drafts. Both are real, but they are not the same job. If someone offers you AI SEO, ask which one they mean.

Google's own answer to "is this still SEO?" is yes. Its guide on AI search says optimising for it is optimising for the search experience, and so still SEO1.

Two meanings, one phrase

We wanted to know what page one actually means by AI SEO, so on 18 October 2026 we checked two UK searches: "AI SEO" and "what is AI SEO". We read the definition each result gives in its title and the snippet Google shows, and sorted the sixteen results into three groups.

The split was almost even. Six results defined AI SEO as getting found in AI answers, seven as using AI tools, and three covered both. Google's own AI Overview for "what is AI SEO" settled it the same way, with a section headed "Two Sides of AI SEO".

So the confusion is not yours. The term is genuinely used for two different things, often by agencies and tool makers with a reason to prefer one meaning.

Meaning one: getting found in AI answers

This is the meaning that changes your strategy. When an AI Overview or an assistant answers a question, it reads several pages and writes one reply. Google says its AI features may run a set of related searches across subtopics and sources to build that reply2. That lets them show a wider set of links than a classic results page2. Your page is either one of those links, or it is not seen at all.

In practice, this side of AI SEO has three parts:

Answering on your own pages. Clear, direct answers to the questions your buyers ask, near the top of each section. This is answer engine optimisation.
Being named across the web. Assistants lean on what other sites say about you, as well as your own pages. This is generative engine optimisation.
The SEO underneath. Google says its AI features rely on the same core ranking and quality systems as normal search1. A page that cannot rank is rarely cited.

Meaning two: using AI tools for SEO

The second meaning is about the workflow, not the result. AI tools can group keywords, draft outlines, summarise competitor pages, suggest titles and spot gaps in a site's content. Used well, they save hours of routine work.

Google does not object to this. Its guidance says generative AI can be useful when researching a topic and when adding structure to original content3. It also warns that AI output can contain inaccuracies, and that checking it before publishing is critical3. That applies to page titles, meta descriptions and image alt text too, because they can appear in search results3.

The line Google draws is volume without value. Its spam policies name using generative AI tools to produce many pages without adding value for users as an example of scaled content abuse4. A tool that helps you write one better page is fine. A tool that writes five hundred thin pages is a risk.

What the AI Overview actually cited

We also looked at where the two AI Overviews took their answers from. Between them they cited twenty two links. Nine of those were YouTube videos, Reddit threads, a LinkedIn post or an Instagram reel, rather than a company website or a publisher's article.

That is a small sample from one day, and the mix changes from search to search. It still makes a point that most AI SEO guides skip. When an AI answer is written, your website is only one of the places it can learn about your subject. Videos, forum threads and posts by real people are read too, which is why the work reaches beyond your own pages.

What changes for your website

Less than many guides suggest. Google says there are no extra requirements to appear in AI Overviews or AI Mode beyond being indexed and eligible to show a snippet2. It also says you do not need new machine readable files, AI text files or special markup to appear in its search1. Nor do you need to break your content into tiny pieces1.

What does change is the bar for content. Google's guide singles out what it calls non-commodity content: material that goes beyond common knowledge, with real expert or first-hand insight1. An AI system can already summarise the generic version of any topic. It needs a reason to cite yours.

For most UK businesses, that means a short list of changes:

Fix what stops pages being found. Indexing, crawl errors, slow templates and duplicate pages. A technical SEO review usually comes first.
Answer the real questions directly. Put a short, plain answer at the top of each section, then the detail. Write it so it still makes sense when quoted on its own.
Add what only you know. Your prices or price ranges, your process, your results, your own tests and photographs. This is the part a competitor's AI tool cannot copy.
Keep your facts the same everywhere. Your services, locations and claims should match on your site, your profiles and directories.
Earn mentions elsewhere. Reviews, trade press, podcasts, guest articles and useful answers in the forums your buyers read.
Check AI output before it goes live. If you use AI tools to draft, a person who knows the subject should edit and fact check every page.

AI SEO and traditional SEO compared

	Traditional SEO	AI SEO (getting found in AI)
The aim	Rank a page in the list of results	Be cited or named in the written answer
Where people see you	A blue link they choose to click	A source link or a mention inside an answer
What counts most	Relevance, quality and links	The same, plus clear answers and first-hand detail
Where the work happens	Mostly your own website	Your website and the wider web
How you measure it	Rankings, clicks and enquiries	Citations, mentions, clicks and enquiries

The two overlap far more than they differ. Our guide to GEO vs SEO goes into where the lines fall.

How to measure AI SEO

Start with what you already have. Google says traffic from its AI features is counted in the normal Performance report in Search Console, under the Web search type2. Google's guide also points to a generative AI performance report in Search Console1. It warns that no third party tool has access to its internal ranking or AI systems1.

Then add two simple checks:

A monthly question list. Write down the questions that matter most to your sales. Each month, search them in Google from the UK and ask them in the main assistants, and note whether you are cited or named.
Ask every new lead. "How did you find us?" picks up the enquiries that no report can see, such as someone who read your name in an AI answer and searched for you directly.

Common questions

Is AI SEO the same as GEO? Close, but not exactly. Many people use AI SEO as an umbrella term for both AEO and GEO. Others use it to mean AI tools for SEO. The label matters less than knowing which job is being done.

Is SEO dead because of AI? No. Google says AI search still depends on its core ranking systems, and that optimising for it is still SEO1. What has changed is the reward for being average, which is smaller than it was.

Will using AI to write content hurt my rankings? Not by itself. Google judges the content, not the tool3. Inaccurate, unedited or mass produced pages are the problem, and those can breach its spam policies4.

Do I need an llms.txt file? Not for Google. It says its search does not use AI text files to decide what appears1. Some other services may read them, so adding one is harmless, but it is not a priority.

Can anyone guarantee AI citations? No. AI answers change from one search to the next, and no outside tool can see inside Google's systems1. Be wary of anyone who promises a place in them.

Where to start

Search the three questions your best customers ask before they buy, in Google and in ChatGPT. Note who is cited, and what those pages have that yours do not. That gap is your first AI SEO plan.

If you would like a second pair of eyes, an SEO audit is the quickest way to find what is holding your site back. Our AI search page explains how we approach the rest.