The AI search glossary.
An AI search glossary in plain English: the GEO terms behind AI visibility, defined so you can talk about them with confidence.
What is GEO (generative engine optimization)?
GEO, or generative engine optimization, is the work of getting your business named in the answers ChatGPT and other AI assistants give when customers ask for a recommendation. Where SEO ranks your pages in a list of links, GEO aims for the answer itself: the few businesses an AI assistant names.
What is AI visibility?
AI visibility is how often, and how prominently, ChatGPT and other AI assistants name your business when customers ask them about what you sell. It is read per question and per assistant, because each one can answer the same question differently.
What is a citation in AI answers?
A citation in an AI answer is a source the assistant links to or draws from when it makes a claim, such as a review site, a Reddit thread or a company page.
GEO vs SEO: what is the difference?
GEO and SEO differ in where you appear: SEO ranks your pages in a list of search results, while GEO gets your business named inside the answers AI assistants give. Both start from a website search engines can read; GEO then puts more weight on what other websites say about you.
What is an AI Overview?
An AI Overview is the AI-generated summary Google shows above the traditional results for some searches, produced by its Gemini models from pages retrieved for that query. It answers the question directly and links to a selection of the sources it drew on.
What is AI share of shelf?
AI share of shelf is the extent to which your brand is one of the few names an AI assistant gives when it answers a buying question. A search results page lists ten links; an AI answer typically recommends a handful of brands, and that short list is the shelf.
What is entity SEO?
Entity SEO is the practice of optimising for how search engines and AI systems understand the things your content is about, such as your brand and its products, rather than the keywords on the page. An entity is any distinct thing a machine can identify and connect to other things, and clear entities are the foundation of accurate AI answers.
What is AI brand monitoring?
AI brand monitoring is the practice of checking what AI assistants such as ChatGPT, Claude, Gemini and Perplexity say about your brand, and tracking how those answers change over time. It treats AI answers as a channel to be watched, the way you would monitor press coverage or search rankings.
What is Google AI Mode?
Google AI Mode is the conversational search experience inside Google Search that answers queries with a full AI-generated response and supports follow-up questions, rather than showing a ranked list of links. Since the Connected Apps rollout began in July 2026, initially in the US, AI Mode answers can also complete tasks, such as adding groceries to an Instacart cart, directly from the response.
What is an AI crawler?
An AI crawler is an automated bot run by an AI company to fetch web pages. The term covers three different jobs: collecting training data for models, building the search indexes behind AI answers, and fetching pages live when a user asks an assistant a question.
What is GPTBot?
GPTBot is OpenAI's web crawler for collecting training data for its GPT models. Blocking it in robots.txt asks OpenAI not to use your content in future training, but it does not affect your visibility in ChatGPT search, which relies on a separate crawler called OAI-SearchBot.
What is Google-Extended?
Google-Extended is a robots.txt control token that tells Google not to use your content for Gemini training or grounding. Blocking it has no effect on your presence in Google Search, AI Overviews or AI Mode.
What is zero-click search?
Zero-click search is a search that ends without a visit to any website: the user reads the answer directly on the results page or inside an AI-generated response. As AI answers handle more complex questions, a growing share of searches finish this way.
What is AI referral traffic?
AI referral traffic is the visits a website receives when someone clicks through from an AI assistant's answer, such as a cited source in ChatGPT or Perplexity. In analytics it usually shows up under referrer hostnames like chatgpt.com and perplexity.ai.
What is prompt coverage?
Prompt coverage is the share of the buyer questions in your category that your AI visibility checks actually test. Two prompts sample a sliver of the question space; meaningful coverage spans the whole buying journey, from unbranded research questions to direct comparisons.
What is AI search?
AI search means getting answers from an AI assistant such as ChatGPT, Claude, Perplexity, or Gemini, which reads multiple sources and synthesises them into a single conversational response instead of returning a ranked page of links.
What is ChatGPT SEO?
ChatGPT SEO is the practice of getting a brand described accurately and recommended in ChatGPT's answers, both when the model replies from its trained memory and when it searches the live web through Bing's index and cited sources.
What is Perplexity SEO?
Perplexity SEO is the practice of optimizing content so Perplexity's answer engine retrieves it during its live web search and cites it as a source, which depends on crawler access, quotable page structure, and presence on domains Perplexity already trusts.
What is share of model?
Share of model is the percentage of an AI assistant's answers about a category that mention or recommend your brand, measured by running a fixed set of prompts repeatedly and counting how often each brand appears.
What is LLM SEO?
LLM SEO is the practice of optimizing a brand's content and web presence so that AI assistants built on large language models, such as ChatGPT, Claude, Gemini, and Perplexity, describe the brand accurately and cite it in their answers. The same work is also called Generative Engine Optimization.
What is a knowledge graph?
A knowledge graph is a structured database of entities, such as companies and products, and the relationships between them; AI systems consult graphs like Google's Knowledge Graph and Wikidata to establish basic facts about a brand before answering.
What is structured data for AI search?
Structured data for AI search is Schema.org markup added to a page's HTML that labels entities, products, FAQs and other facts in machine-readable form, helping AI crawlers and answer engines interpret content, though most assistants still read the rendered text itself.
What is an AI visibility audit?
An AI visibility audit is a structured check of how AI assistants such as ChatGPT, Claude and Perplexity answer the prompts buyers actually ask, testing whether a brand appears in those answers and whether the description of it is accurate.
More terms
What is AEO (answer engine optimization)?
AEO, or Answer Engine Optimization, is the practice of structuring content so answer engines can extract a clean, correct answer from it and cite it.
AEO overlaps heavily with GEO. It focuses on the on-page side: answer-first writing, self-contained sections, clear headings phrased as questions, and structured data.
AEO vs GEO vs SEO: what is the difference?
AEO and GEO both mean getting your business into the answers that search engines and AI assistants give, while SEO means getting your pages ranked in search results. In day-to-day work AEO and GEO are the same job.
AEO stands for answer engine optimization and GEO for generative engine optimization, and Google describes both as terms for work on visibility in AI search. GEO was named in a 2023 research paper, GEO: Generative Engine Optimization, for AI search tools that write an answer by summarising several sources.
What is a prompt in GEO?
A prompt in GEO is the question or request a user puts to an AI assistant, and it replaces the keyword as the basic unit of visibility tracking. Rather than asking where you rank for a term, GEO asks which businesses the assistant names and cites when it answers a given prompt.
Prompts differ from keywords in length and intent. A keyword is a compressed fragment typed into a search box, such as "accounting software uk". A prompt is a full request with context attached: "what's the best accounting software for a UK sole trader who does everything on their phone?" The intent that SEO tools infer from keywords is stated outright in a prompt, which is why AI answers can be so specific about who they recommend.
What is brand grounding?
Brand grounding is how an AI assistant anchors the claims it makes about a brand in retrieved web pages rather than in its training data alone. A grounded answer about your pricing or features is drawn from live sources the engine can cite; an ungrounded one is reconstructed from whatever the model absorbed during training.
The mechanics follow a retrieve-then-generate pattern. When an engine judges that a query needs current information, it runs a web search, reads a handful of pages and composes its answer from them, usually attaching citations. ChatGPT with browsing, Perplexity, Copilot, Gemini and Claude with web search all work broadly this way, though each engine's triggering rules and source selection differ and none of the internals are fully public.
What is retrieval-augmented generation (RAG)?
Retrieval-augmented generation (RAG) is a technique where an AI system fetches relevant documents, often from a live web index, and passes them to a language model to ground its answer. Instead of relying only on what the model memorised during training, the answer is built from content retrieved at the moment you ask. For marketers, it means an AI assistant's description of your brand can come straight from web pages, including yours.
The process has two stages. When you ask a question, the system first searches an index for relevant material, then feeds the best passages to the model alongside your question, and the model writes its answer with those passages in front of it. That is why grounded AI answers often include citations: the sources are not decoration, they are the raw material the answer was assembled from.
What is an AI citation gap?
An AI citation gap is the difference between how often AI assistants cite your content in relevant answers and how often they cite your competitors'. If ChatGPT or Perplexity answers your category's buying questions with links to rival sites and none to yours, you have a citation gap.
The gap matters because AI answers have far fewer slots than a search results page. A grounded answer typically cites a handful of sources rather than ten links, so the sources that are selected carry disproportionate weight. When a buyer asks an assistant to compare options, the brands whose pages and coverage get cited are effectively the shortlist, and everyone else is absent from the conversation.
What is llms.txt?
llms.txt is a proposed standard: a markdown file placed at the root of a website that gives large language models a curated summary of the site and links to its most important content. It was proposed by Jeremy Howard of Answer.AI in September 2024 as a way to help AI systems use a site's key pages without wading through navigation, scripts and boilerplate HTML.
The format is deliberately simple. The file lives at yourdomain.com/llms.txt and contains a title, a short summary in a blockquote, and lists of links to key pages with one-line descriptions of each. A companion convention, llms-full.txt, goes further and includes the full text of the site's important content in a single file. The logic is that model context windows are finite and HTML is noisy, so a curated, machine-readable map lets an AI system spend its context on your actual content.
What is an answer engine?
An answer engine is a search system that responds to a query with a single synthesised answer instead of a ranked list of links. Systems such as ChatGPT, Perplexity, Google AI Mode and Claude read multiple sources and compose one response, so a brand is either included in the answer or absent from it.
A traditional search engine takes a query and returns a ranked list of pages, then leaves the choosing to you. An answer engine does the reading itself. ChatGPT with search enabled, Perplexity, Google AI Mode, Microsoft Copilot and Claude all work this way: they retrieve a set of sources and compose a single written response from them, usually citing only a handful of the pages they drew on.
What is content chunking?
Content chunking is the practice of structuring a page into self-contained sections that an AI engine can retrieve and quote on their own. Because answer engines typically pull passages rather than whole pages, each section needs to answer its question fully and name your brand without relying on the surrounding text.
When an answer engine responds to a question, it rarely reads your whole page. Retrieval systems typically split pages into passages and score each one against the query, then hand the best few to the model that writes the answer. That means the unit of competition is not the page but the section. A well-known site with the answer buried in paragraph twelve can lose to a smaller site whose relevant section stands cleanly on its own.
What is E-E-A-T in AI search?
E-E-A-T stands for experience, expertise, authoritativeness and trustworthiness, the framework Google uses to judge content quality. AI answer engines have no direct equivalent, but they inherit its effects because they retrieve and cite sources from indexes already shaped by E-E-A-T-driven ranking.
E-E-A-T stands for experience, expertise, authoritativeness and trustworthiness. It comes from Google's Search Quality Rater Guidelines, where human reviewers use it to judge whether a page deserves its ranking, and Google says rater feedback helps it tune ranking systems towards the qualities those reviewers reward. It has never been a single score, even inside Google. The second E, experience, was added in December 2022 to favour content written by people who have actually used the product or lived through the situation they describe.
What is a brand hallucination?
A brand hallucination is when an AI assistant presents false information about a company as fact, such as wrong pricing or a claim that the company has shut down, with nothing in the answer to signal uncertainty.
Brand hallucinations tend to be concrete and checkable: a price quoted from a plan you retired in 2024, or a confident claim that a feature exists when it never shipped. In our July 2026 study, 27 of 30 brands were materially misdescribed by an AI assistant answering from memory, and 13 of 30 were still misdescribed when the assistant had live web search. Because the answer arrives in fluent, assured prose, buyers rarely question it.
What is agentic commerce?
Agentic commerce is the buying of goods and services by an AI agent acting for a human: the person states what they need, and the agent finds suitable options and completes the purchase, often without the buyer visiting any brand website.
A typical agentic purchase starts with a one-line request. A buyer asks an assistant for a waterproof hiking jacket under 200 dollars; the agent reads reviews and comparison pages, then settles on a product. On supported platforms it can pay too: OpenAI's Instant Checkout, built with Stripe on the Agentic Commerce Protocol, already completes orders inside ChatGPT for participating Etsy and Shopify sellers, and Google's Agent Payments Protocol covers similar ground.
See where you stand in AI search, and fix it.
Run a free check in under a minute. No card, no login to see your score.