AI Search Glossary: 65 AI SEO Terms Defined, With Sources

These are the 65 AI search terms you'll hear in pitches and meetings, each defined in one to three plain sentences, with a source link for all but the made-up ones. Of the 65, 47 are sourced straight to Google, OpenAI, Anthropic, Microsoft or Perplexity, and the last seven are terms to stop using.

The conversion terms that apply once a visitor lands are in our CRO glossary.

What does each AI search term mean, at a glance?

Term	What it means	Use it?
AI search engines and features
AI search	AI writes the answer from fetched pages	Yes
AI Overviews	Google's AI summary in normal results	Yes
AI Mode	Google's separate AI search	Yes
ChatGPT search	ChatGPT answering from the web	Yes
Perplexity	AI search engine that cites sources	Yes
Microsoft Copilot	Microsoft's AI assistant, using Bing's index	Yes
Gemini	Google's AI models and app	Yes
Claude search	Claude's cited web search	Yes
AI agents (agentic search)	AI doing tasks on websites for you	Yes
Getting cited
GEO	Work to get cited in AI answers	Yes
AEO	Another name for the same work	Yes
LLMO	A rarer synonym for GEO	Niche
AI SEO	The everyday name. Google calls it SEO	Yes
Citation	A link to your page as a source	Yes
Mention	An AI answer naming your brand	Yes
Citation share	Your percentage of a search's citations	Yes
Share of model	Your share of AI brand mentions	Niche
Zero-click search	A search with no click	Yes
Answer-first opening	The answer in the first sentence or two	Yes
BLUF	"Bottom line up front", the Army's answer-first	Yes
Information gain	What a page adds that's new	Yes
Citability	How quotable a page is to AI	Niche
How AI answers are built
RAG	Look things up, then write	Yes
Grounding	Tying answers to real sources	Yes
Query fan-out	One question split into several searches	Yes
Passage or chunk	A page section fetched alone	Yes
Embeddings and vector search	Matching text by meaning	Niche
Entity	A distinct real-world thing software recognises	Yes
Knowledge graph	A database of entities and how they relate	Yes
Hallucination	An AI stating something false	Yes
Non-determinism	A different answer each time	Yes
Crawlers and access
AI crawler	A bot sent by an AI company	Yes
Training, search and user-triggered bots	The three jobs AI bots do	Yes
GPTBot	OpenAI's training crawler	Yes
OAI-SearchBot	OpenAI's crawler for ChatGPT search	Yes
ChatGPT-User	Fetches pages for ChatGPT users	Yes
ClaudeBot	Anthropic's training crawler	Yes
Claude-SearchBot	Anthropic's crawler for Claude's search	Yes
Claude-User	Fetches pages for Claude users	Yes
PerplexityBot	Perplexity's search crawler	Yes
Perplexity-User	Fetches pages for Perplexity users	Yes
Google-Extended	A robots.txt setting for Gemini training	Yes
Google-Agent	Fetcher for AI agents on Google	Yes
Bingbot	The crawler behind Bing and Copilot	Yes
robots.txt	Tells crawlers what they may visit	Yes
Server-side rendering	Content readable without JavaScript	Yes
IndexNow	Tells Bing and others a page changed	Yes
Measuring
AI visibility tracking	Counting mentions and citations across prompts	Yes
Bing AI Performance report	Your Copilot citations, free from Bing	Yes
Search Console Generative AI performance report	Your impressions in AI Overviews and AI Mode	Yes
Search generative AI control	On-off switch for Google's AI features	Yes
AI referral traffic	Visits from clicks in AI answers	Yes
GA4 AI Assistant channel	GA4's label for AI assistant visits	Yes
Trust and structure
E-E-A-T	Experience, expertise, authoritativeness and trust	Yes
Structured data (schema markup)	Code that labels page content	Yes
JSON-LD	Google's preferred structured data format	Yes
Schema and AI citations	No citation lift in Ahrefs' test	Yes
Named author	A real person credited on the page	Yes
Terms to stop using
LSI keywords	Google says there's no such thing	Retire
AMP for Top Stories	Not required since 2021	Retire
llms.txt ("you need one")	A proposed AI file Google ignores	Retire
SEM as "semantic search"	SEM means search engine marketing	Retire
"Chunk your content for AI"	Google says there's no need	Retire
Buying mentions	Google says it helps less than it seems	Retire
CIT, MEN, GRD, HAL, MEM	Made-up acronyms we found nobody using	Retire

"Yes" means worth knowing. "Niche" means real but rare. "Retire" means wrong, out of date or made up.

Which AI search engines and features should you know?

The ones to know are Google's AI Overviews, AI Mode and Gemini, OpenAI's ChatGPT search, Microsoft Copilot, Perplexity and Anthropic's Claude. Each answers questions from the live web and links to some of its sources.

What is AI search?

AI search is any search where an AI writes the answer from web pages it has looked up, and can link to them. Google says its versions are "rooted in our core Search ranking and quality systems" (Google, 2026).

What are AI Overviews?

AI Overviews are Google's AI-written summaries in normal search results, with links to supporting pages. To be one of those links, a page must be indexed and snippet-eligible, and Google says no special files or schema are needed (Google, 2025).

What is AI Mode?

AI Mode is Google's separate AI search experience for questions needing "further exploration, reasoning, or complex comparisons" (Google, 2025). Behind each question it runs several related searches, which Google calls query fan-out.

What is ChatGPT search?

ChatGPT search lets ChatGPT "search the web to answer questions with current information and links to relevant sources", and OpenAI says it "sometimes partners with other search providers" (OpenAI, 2026). Our ChatGPT SEO playbook covers how pages get picked.

What is Perplexity?

Perplexity is an AI search engine that writes a conversational answer to each question, and "Each response includes citations and links to original sources" (Perplexity Help Center). Perplexity runs its own search index, which it says "tracks over 200 billion unique URLs" (Perplexity, 2025). Our Perplexity SEO guide covers getting cited.

What is Microsoft Copilot?

Copilot is Microsoft's AI assistant. Bing says Copilot and Bing search "rely on the same core crawling, indexing, and ranking foundation as traditional search" (Bing Webmaster Guidelines), which makes Bingbot the crawler that matters, as our Copilot SEO playbook explains.

What is Gemini?

Gemini is Google's family of AI models and its assistant app, which replaced Bard in February 2024 (Google, 2024). Our Claude and Gemini SEO guide covers both engines.

What is Claude search?

Claude search is the web search tool in Anthropic's Claude assistant, launched on 20 March 2025 to "deliver up-to-date, cited information in conversational responses" (Anthropic, 2025).

What are AI agents (agentic search)?

AI agents are AI systems that do tasks for a person on websites, such as "booking a reservation or comparing product specifications" (Google, 2026). We found no vendor defining "agentic search" separately.

What does getting cited by AI mean?

Getting cited means an AI answer links to your page as a source. A mention, where the answer names your brand, is a separate measure. GEO, AEO, LLMO and AI SEO are four names for the work of earning both.

What is GEO (generative engine optimisation)?

GEO is the work of making your content more likely to be shown and cited in AI-written answers. The name comes from a 2023 research paper by Aggarwal and colleagues, and Bing's Webmaster Guidelines now define it too.

What is AEO (answer engine optimisation)?

AEO is another common name for getting your content used in AI answers. Google treats it as SEO (Google, 2026), and even the expansion isn't settled: Microsoft's January 2026 guide spells it "Answer/Agentic Engine Optimization" (Search Engine Journal, 2026).

What is LLMO?

LLMO (large language model optimisation) is a rarer name for GEO. Conductor's glossary calls it "essentially synonymous" with GEO (Conductor).

What is AI SEO?

AI SEO is the everyday name for SEO aimed at AI answers, which Google says is still SEO:

"From Google Search's perspective, optimizing for generative AI search is optimizing for the search experience, and thus still SEO." (Google, 2026)

Our AI SEO strategy guide covers the work.

What is a citation in AI search?

A citation is a link to your page shown as a source in an AI-written answer. It's what Bing's AI Performance report counts: "citations that are displayed as sources in AI-generated answers" (Bing, February 2026). We compared what ChatGPT actually cites across six studies.

What is a mention in AI search?

A mention is an AI answer naming your brand or product, with or without a link. In Kevin Indig's study of Semrush ChatGPT data, citations and mentions were "only weakly related, with a slightly negative correlation of -0.229" (Search Engine Journal, 2026).

Citation share is Bing's measure of "the percentage of citations attributed to your site out of all citations shown across all sites" (Bing, June 2026). "Share of voice in AI" is a looser phrase with no standard definition, so ask what any tool counts.

Share of model is a brand's share of all the brand mentions AI models give in its category. Tom Roach of the agency Jellyfish set out the measure in Marketing Week in July 2024.

What is a zero-click search?

A zero-click search is one that ends without a click on any result (SparkToro, 2024). That includes people who got their answer, gave up or searched again, so AI isn't the only cause.

What is an answer-first opening?

An answer-first opening puts a direct answer to the page's main question in its first sentence or two. For AI answers, Microsoft recommends "One- to two-sentence responses that directly address a question" (Microsoft Advertising, 2025). Our guide to the first 50 words shows how.

What is BLUF?

BLUF, "bottom line up front", is the US Army rule of putting the main point first (US Army, AR 25-50), and an older name for an answer-first opening.

What is information gain?

Information gain is the new information a page adds beyond what the reader has already seen, an idea from a Google patent granted in June 2022. A patent doesn't prove Google ranks on it, though Google's AI guide does ask for "non-commodity content" (Google, 2026).

What is citability?

Citability is how likely a page is to be quoted or cited in AI answers, judged from the page itself. No primary source defines it. The observational GEO-16 study of 70 prompts found "overall page quality is a strong predictor of citation". Rubric, the tool we make, estimates citability and never counts citations.

How are AI answers built?

AI search engines look things up before they write, which Google calls RAG or grounding. One question can trigger several searches, called query fan-out, and the finished answer can still contain errors, called hallucinations.

What is retrieval-augmented generation (RAG)?

RAG is when an AI looks up relevant documents first, then writes its answer from what it found. The term comes from a 2020 paper by Lewis and colleagues, and Google calls RAG "A technique (also known as grounding)" (Google, 2026). Our AI retrieval guide covers your side.

What is grounding in AI search?

Grounding ties an AI's answer to real sources, such as live web pages, so it can show where its facts came from (Google AI for Developers). Bing calls the searches its AI runs "grounding queries" (Bing, June 2026), and our grounding guide shows how to be that source.

What is query fan-out?

Query fan-out is when an AI search splits your question into several searches and runs them at once. Google defines it as "A set of concurrent, related queries generated by the model" (Google, 2026) and publishes no count. It warns that pages made for every fan-out query can breach its scaled content abuse policy, so answer the follow-ups that matter.

What is a passage, or chunk?

A chunk, or passage, is a short section of a document that an AI retrieval system stores and fetches on its own (Anthropic, 2024). Chunk sizes are settings each developer picks, so ignore anyone quoting one as an engine rule.

What are embeddings and vector search?

Embeddings turn text into lists of numbers that stand for its meaning, and vector search uses them to match a question to passages even when the words differ (OpenAI).

What is an entity in SEO?

An entity is a distinct real-world thing, such as a person, business, place or product, that software can recognise (Google Cloud). Using one business name everywhere, with sameAs links in your schema, helps engines match you to one entity.

What is a knowledge graph?

A knowledge graph is a database of real-world things and how they relate. Google launched its Knowledge Graph in May 2012 to understand "real-world entities and their relationships to one another" (Google, 2012).

What is an AI hallucination?

A hallucination is an AI stating something false as if it were true, or in OpenAI's words, "plausible but false statements generated by language models" (OpenAI, 2025). Grounding is meant to reduce them (Google AI for Developers), but it doesn't remove them.

What is non-determinism in AI answers?

Non-determinism means an AI can give a different answer each time you ask the same question. SparkToro's January 2026 study of 2,961 runs found less than a 1 in 100 chance that ChatGPT or Google's AI would give "the same list of brands in any two responses" (SparkToro, 2026). One check of one answer proves little.

Which AI crawlers visit your site, and what controls them?

Each AI company runs separate bots for separate jobs. Block a training crawler and you stay in AI search, but block a search crawler and you can drop out. Our guide to getting indexed by AI search engines shows how to check.

What is an AI crawler?

An AI crawler is a bot run by an AI company that visits web pages to train models, build a search index or answer a user's request (OpenAI).

What types of AI crawler are there?

There are three types: training crawlers collect pages to build models, search crawlers build the index behind cited answers, and user-triggered fetchers open a page when someone's question needs it. Of its own bots, OpenAI says "Each setting is independent of the others" (OpenAI).

Company	Training	Search index	User-triggered
OpenAI	GPTBot	OAI-SearchBot	ChatGPT-User
Anthropic	ClaudeBot	Claude-SearchBot	Claude-User
Perplexity	None (says it doesn't train models)	PerplexityBot	Perplexity-User
Google	Google-Extended (a robots.txt setting only)	Googlebot	Google-Agent and others

What is GPTBot?

GPTBot is OpenAI's training crawler. OpenAI says "Disallowing GPTBot indicates a site's content should not be used in training generative AI foundation models" (OpenAI), and blocking it doesn't affect ChatGPT search, which uses OAI-SearchBot.

What is OAI-SearchBot?

OAI-SearchBot is OpenAI's search crawler, used "to surface websites in search results in ChatGPT's search features". To appear in ChatGPT search, allow it and OpenAI's published IP ranges, and give robots.txt changes about 24 hours to apply (OpenAI).

What is ChatGPT-User?

ChatGPT-User is the agent ChatGPT uses to open a page when a user's request needs it. OpenAI says it doesn't crawl the web automatically, and because a user starts the action, "robots.txt rules may not apply" (OpenAI).

What is ClaudeBot?

ClaudeBot is Anthropic's training crawler, and restricting it tells Anthropic to keep your future pages out of its model training (Anthropic, 2026).

What is Claude-SearchBot?

Claude-SearchBot is Anthropic's search crawler, which indexes pages so Claude can find and cite them. Anthropic says blocking it "prevents our system from indexing your content for search optimization" and may cut your Claude search visibility (Anthropic, 2026).

What is Claude-User?

Claude-User is the agent Claude uses to fetch a page when a person's question needs it. Anthropic says it honours robots.txt, so blocking it may cut your visibility in Claude's searches (Anthropic, 2026). OpenAI and Perplexity say their user fetchers may not follow it.

What is PerplexityBot?

PerplexityBot is Perplexity's search crawler, which Perplexity says "is not used to crawl content for AI foundation models" (Perplexity).

What is Perplexity-User?

Perplexity-User is the agent Perplexity uses to visit a page when a user asks a question, and "this fetcher generally ignores robots.txt rules" because the user requested the fetch (Perplexity).

What is Google-Extended?

Google-Extended is a robots.txt setting that controls whether Google may use your pages to train Gemini and ground Gemini app answers. Google says it "does not impact a site's inclusion in Google Search" (Google, 2026), and it doesn't control AI Overviews or AI Mode, which follow your Googlebot rules and the Search generative AI control.

What is Google-Agent?

Google-Agent is the fetcher AI agents on Google's infrastructure use to browse websites and "perform actions upon user request", and Google says its user-triggered fetchers "generally ignore robots.txt rules" (Google, 2026).

What is Bingbot?

Bingbot is Microsoft's main crawler, "our standard crawler" in Bing's words, collecting the pages behind Bing search and Copilot (Bing Webmaster Tools). Bing lists no separate Copilot crawler, and its guidelines list blocking Bingbot as something to avoid.

What is robots.txt?

Robots.txt is a plain text file at the root of your site that tells crawlers which parts they may visit. RFC 9309 says "These rules are not a form of access authorization", so it's a request, and your firewall can still block bots it allows. It only governs fetching. Google's Gary Illyes said in October 2026 that robots.txt "doesn't dictate what a system does with the data after it has already crawled it" (Search Engine Journal).

What is server-side rendering?

Server-side rendering sends finished page content in the HTML, while JavaScript rendering builds it in the browser. Vercel's December 2024 analysis found "none of the major AI crawlers currently render JavaScript", so check your pages' raw HTML.

What is IndexNow?

IndexNow is "a simple ping so that search engines know that a URL and its content has been added, updated, or deleted" (IndexNow.org). Amazon, Bing, Naver, Seznam.cz, Yandex and Yep take part, and Google doesn't (IndexNow.org).

How do you measure AI search?

Bing, Google Search Console and GA4 give you free first-party data on AI search, and repeated prompt tests cover the engines that publish none. No single tool sees every AI answer.

What is AI visibility tracking (prompt tracking)?

AI visibility tracking means running the same prompts through AI tools many times and recording how often your brand is mentioned or cited. SparkToro calls percentages across prompts run dozens of times "a reasonable metric" (SparkToro, 2026). Google's warning applies to every AI SEO monitoring tool:

"No third-party tool has access to our internal ranking or AI systems." (Google, 2026)

What is Bing's AI Performance report?

Bing's AI Performance report is a free Bing Webmaster Tools report showing how often Copilot and Bing's AI answers cite your pages, and for which searches (Bing Webmaster Tools). It doesn't cover ChatGPT, its query data is sampled, and citation share arrived in June 2026.

What is Search Console's Generative AI performance report?

Search Console's Generative AI performance report shows how often links to your site appeared in Google's AI Overviews and AI Mode (Search Console Help). It counts impressions only, with no click data, and rolled out worldwide on 31 August 2026.

What is the Search generative AI control?

The Search generative AI control is a Search Console setting that includes or excludes your site from AI Overviews, AI Mode and Discover's AI features (Search Console Help). It's on by default, sites must be included to be eligible (Google, 2026), and it doesn't affect AI training, which is Google-Extended's job.

What is AI referral traffic?

AI referral traffic is visits that arrive when someone clicks a link in an AI assistant's answer. GA4 counts clicks from Google's AI Overviews and AI Mode as Organic Search, so no single channel shows all of it (Google Analytics Help).

What is GA4's AI Assistant channel?

GA4's AI Assistant channel, added on 13 May 2026, groups visits "from sources like ChatGPT, Gemini, Deepseek, Copilot, or Grok" (Google Analytics Help).

Which trust and structure terms still matter for AI search?

Named authors and structured data still matter, for readers and rich results more than for AI citations. Google says E-E-A-T isn't a ranking factor and structured data isn't required for its AI features.

What is E-E-A-T?

E-E-A-T stands for experience, expertise, authoritativeness and trust, the qualities Google's human quality raters look for in a page (Google, 2025). Google also says "E-E-A-T itself isn't a specific ranking factor" (Google, 2026).

What is structured data (schema markup)?

Structured data, or schema markup, is code that labels a page's content, such as a price or an author, for search engines (Google, 2025). Google says it "isn't required for generative AI search" (Google, 2026), though it still makes pages eligible for rich results.

What is JSON-LD?

JSON-LD is a code format for structured data that sits in a script block on the page (W3C, 2020), and it's the format Google recommends (Google, 2025). If JavaScript adds it after the page loads, AI crawlers that don't run JavaScript can miss it.

What's the evidence on schema and AI citations?

Adding schema didn't get pages cited more in Ahrefs' test of 1,885 pages that added it, matched against 4,000 control pages from August 2025 to March 2026:

"Adding schema produced no major uplift in citations on any platform." (Ahrefs, 2026)

AI Overview citations fell 4.6%, and the AI Mode (+2.4%) and ChatGPT (+2.2%) changes were too small to tell apart from zero (Ahrefs, 2026).

What is a named author?

A named author is a real, identified person credited on a page with a byline. Google asks for "accurate authorship information, such as bylines" where readers expect them (Google, 2026).

Missing bylines were the most common failure in GoGoChimp's 7 October 2026 benchmark re-crawl: 198 of 242 sites (82%) left a named author off more than half the pages that needed one.

Which AI SEO terms should you stop using?

Stop using LSI keywords, AMP for Top Stories, "you need llms.txt", SEM as "semantic search", "chunk your content for AI", bought mentions and made-up acronyms like CIT and GRD. Google itself contradicts five of these seven. If one turns up in a sales pitch, ask where it comes from.

Are LSI keywords real?

LSI keywords are related words that supposedly help a page rank, but Google's John Mueller said in July 2019, "There's no such thing as LSI keywords" (Search Engine Roundtable, 2019). Google's 2026 AI guide says "AI systems can understand synonyms" (Google, 2026), so you don't need every keyword variation.

Do you still need AMP for Top Stories?

You haven't needed AMP (Accelerated Mobile Pages) for Top Stories since 2021. Google said in April 2021 that "using the AMP format is no longer required" for the carousel (Google, 2021).

Do you need llms.txt for AI search?

You don't need llms.txt for Google, whose AI guide says Google Search ignores "AI text files" and that adding one "will neither harm nor help" (Google, 2026). llms.txt is a proposed Markdown file that summarises a site for AI tools, suggested by Jeremy Howard in September 2024 (llmstxt.org). "LLMs" on its own just means large language models.

Does SEM mean semantic search?

SEM stands for search engine marketing, meaning traffic won through better rankings or paid search ads (Dictionary.com). In everyday use it can mean paid ads alone, and we found no source using it for semantic search.

Should you chunk your content for AI?

Google says you don't need to: "There's no requirement to break your content into tiny pieces for AI to better understand it" (Google, 2026). Write clear sections for readers and leave the chunking to the engines.

Should you buy mentions to get into AI answers?

Buying mentions is a poor bet. Google's AI guide warns that "seeking inauthentic 'mentions' across the web isn't as helpful as it might seem" (Google, 2026).

What do CIT, MEN, GRD, HAL and MEM mean?

CIT, MEN, GRD, HAL and MEM are made-up abbreviations from social media graphics, the first four standing in for citation, mention, grounding and hallucination. We found no search engine, AI company or research paper using them. They're the labels on the fuse box in a rented flat: official-looking, and no help to anyone trying to use them.

Questions about AI search terms

What's the difference between GEO, AEO and SEO?

There's very little difference in practice, because Google treats GEO and AEO as SEO (Google, 2026). The acronym an agency uses tells you more about its marketing than its methods, so judge the work by what it measures. Our GEO, SEO, AEO and AIO comparison draws the lines.

Does schema markup get you cited by AI?

Schema doesn't get you cited on its own (Ahrefs, 2026). Add it anyway if you sell products or publish articles, because Product and Article markup labels prices, stock and authors and makes pages eligible for rich results. Just don't pay for it as a citation fix.

Do I need an llms.txt file?

You don't need one for Google, which ignores the file (Google, 2026). If you already have one, leave it, since it does no harm. If you don't, spend the afternoon on robots.txt and your firewall instead: blocking OAI-SearchBot can keep you out of ChatGPT search (OpenAI).

What's the difference between a citation and a mention?

A citation links to your page, and a mention names your brand. They move separately: in Kevin Indig's ChatGPT study, the most-cited domain was the most-mentioned brand only 20.8% of the time (Search Engine Journal, 2026). Bing's AI Performance report counts citations, so tracking mentions takes repeated prompt tests. Report them in separate columns.

What does an AI crawler do, and should you block one?

It fetches your pages for an AI company. To stay out of training but visible in AI search, disallow GPTBot and ClaudeBot and allow OAI-SearchBot, Claude-SearchBot, PerplexityBot, Bingbot and Googlebot (OpenAI). Disallowing Google-Extended keeps you out of Gemini training and Gemini app grounding alike.

How do you measure AI visibility?

Start with the free reports: Bing's AI Performance report for Copilot citations, Search Console for AI Overview and AI Mode impressions, and GA4's AI Assistant channel for visits. For ChatGPT, Claude and Perplexity, run a fixed prompt set dozens of times a month and report how often you're cited or mentioned. Our AI visibility tracking guide sets it up.

Chris McCarron founded GoGoChimp, a Glasgow CRO agency, in 2013 and has 13 years of hands-on CRO experience.

References

Google. (2026, last updated 10 July). Google's guide to optimizing for generative AI features on Google Search. Google Search Central. https://developers.google.com/search/docs/fundamentals/ai-optimization-guide
Google. (2025, last updated 10 December). AI features and your website. Google Search Central. https://developers.google.com/search/docs/appearance/ai-features
OpenAI. (2026, last updated 21 August). Searching the web with ChatGPT. OpenAI Help Center. https://help.openai.com/en/articles/9237897-searching-the-web-with-chatgpt
Perplexity. (n.d., accessed 15 October 2026). What is Perplexity? Perplexity Help Center. https://www.perplexity.ai/help-center/en/articles/10352155-what-is-perplexity
Perplexity Research. (2025, 25 September). Architecting and Evaluating an AI-First Search API. Perplexity blog. https://www.perplexity.ai/hub/blog/architecting-and-evaluating-an-ai-first-search-api
Microsoft. (n.d., accessed 15 October 2026). Bing Webmaster Guidelines. Bing Webmaster Tools. https://www.bing.com/webmasters/help/webmaster-guidelines-30fba23a
Google. (2024, 8 February). Bard becomes Gemini: Try Ultra 1.0 and a new mobile app today. The Keyword. https://blog.google/products-and-platforms/products/gemini/bard-gemini-advanced-app/
Anthropic. (2025, 20 March). Claude can now search the web. Claude blog. https://claude.com/resources/articles/web-search
Aggarwal, P., et al. (2023, 16 November, revised 28 June 2024). GEO: Generative Engine Optimization. arXiv 2311.09735 (KDD 2024). https://arxiv.org/abs/2311.09735
Montti, R. (2026, 22 January). Microsoft's Guide To Winning In AEO & GEO. Search Engine Journal. https://www.searchenginejournal.com/a-breakdown-of-microsofts-guide-to-aeo-geo/565651/
Conductor. (n.d., accessed 15 October 2026). LLMO (Large Language Model Optimization). Conductor Academy glossary. https://www.conductor.com/academy/glossary/llmo-large-language-model-optimization/
Madhavan, K., Merchant, M., Canel, F., & Nigam, S. (2026, 10 February). Introducing AI Performance in Bing Webmaster Tools (Public Preview). Bing Webmaster Blog. https://blogs.bing.com/webmaster/February-2026/Introducing-AI-Performance-in-Bing-Webmaster-Tools-Public-Preview
Jarboe, G. (2026, 29 July). 89% of AI search demand has no clear owner: use this before the window closes (reporting Kevin Indig's analysis of Semrush ChatGPT data). Search Engine Journal. https://www.searchenginejournal.com/89-of-ai-search-demand-has-no-clear-owner-use-this-before-the-window-closes/583255/
Madhavan, K., Merchant, M., Nigam, S., & Shah, T. (2026, 16 June). New AI Visibility Insights in Bing Webmaster Tools: Intents, Topics, Citation Share, Compare. Bing Search Blog. https://blogs.bing.com/search/2026/6/New-AI-Visibility-Insights-in-Bing-Webmaster-Tools-Intents-Topics-Citation-Share-Compare/
Roach, T. (2024, 11 July). 'Share of model' is the new marketing measure for the AI era. Marketing Week. https://www.marketingweek.com/tom-roach-share-of-model-ai-era/
Fishkin, R. (2024, 2 July). 2024 Zero-Click Search Study: For every 1,000 US Google searches, only 374 clicks go to the open web. SparkToro. https://sparktoro.com/blog/2024-zero-click-search-study-for-every-1000-us-google-searches-only-374-clicks-go-to-the-open-web-in-the-eu-its-360/
Madhavan, K. (2025, 8 October). Optimizing your content for inclusion in AI search answers. Microsoft Advertising blog. https://about.ads.microsoft.com/en/blog/post/october-2025/optimizing-your-content-for-inclusion-in-ai-search-answers
US Department of the Army. (2020, 10 October, administrative revision 4 October 2024). Army Regulation 25-50: Preparing and Managing Correspondence, para 1-38b. Copy hosted by the State of Maine. https://www.maine.gov/dvem/policies/documents/AR%2025-50%20%2810%20October%202020%29.pdf
Carbune, V., & Gonnet Anders, P. (2022, 7 June). Contextual estimation of link information gain. US patent 11354342 B2, Google LLC. https://patents.google.com/patent/US11354342B2/en
Kumar, A., & Palkhouski, L. (2025, 13 September). AI Answer Engine Citation Behavior: An Empirical Analysis of the GEO-16 Framework. arXiv 2509.10762. https://arxiv.org/abs/2509.10762
Lewis, P., et al. (2020, 22 May). Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. arXiv 2005.11401 (NeurIPS 2020). https://arxiv.org/abs/2005.11401
Google. (2026, last updated 13 October). Grounding with Google Search. Gemini API docs, Google AI for Developers. https://ai.google.dev/gemini-api/docs/google-search
Anthropic. (2024, 19 September). Introducing Contextual Retrieval. https://www.anthropic.com/engineering/contextual-retrieval
OpenAI. (n.d., accessed 15 October 2026). Vector embeddings. OpenAI API docs. https://developers.openai.com/api/docs/guides/embeddings
Google Cloud. (2026, last updated 13 October). Analyzing entities. Cloud Natural Language API docs. https://docs.cloud.google.com/natural-language/docs/analyzing-entities
Singhal, A. (2012, 16 May). Introducing the Knowledge Graph. Google. https://blog.google/products-and-platforms/products/search/introducing-knowledge-graph-things-not/
OpenAI. (2025, 5 September). Why language models hallucinate. https://openai.com/index/why-language-models-hallucinate/
Fishkin, R., with O'Donnell, P. (2026, 28 January). New research: AIs are highly inconsistent when recommending brands or products. SparkToro. https://sparktoro.com/blog/new-research-ais-are-highly-inconsistent-when-recommending-brands-or-products-marketers-should-take-care-when-tracking-ai-visibility/
OpenAI. (n.d., accessed 15 October 2026). Overview of OpenAI crawlers. OpenAI API docs. https://developers.openai.com/api/docs/bots
Anthropic. (2026, last updated 7 April). Does Anthropic crawl data from the web, and how can site owners block the crawler? Claude Help Center. https://support.claude.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler
Perplexity. (n.d., accessed 15 October 2026). Perplexity crawlers. Perplexity Docs. https://docs.perplexity.ai/docs/resources/perplexity-crawlers
Google. (2026, last updated 14 July). Google's common crawlers. Google Search Central. https://developers.google.com/crawling/docs/crawlers-fetchers/google-common-crawlers
Google. (2026, last updated 19 August). Google's user-triggered fetchers. Google Search Central. https://developers.google.com/crawling/docs/crawlers-fetchers/google-user-triggered-fetchers
Microsoft. (n.d., accessed 15 October 2026). Which crawlers does Bing use? Bing Webmaster Tools. https://www.bing.com/webmasters/help/which-crawlers-does-bing-use-8c184ec0
Koster, M., Illyes, G., Zeller, H., & Sassman, L. (2022, September). RFC 9309: Robots Exclusion Protocol. IETF. https://www.rfc-editor.org/rfc/rfc9309.html
Howard, J. (2024, 3 September, page modified 10 August 2026). The /llms.txt file. llmstxt.org. https://llmstxt.org/
Zecchini, G., Moore, A. A., Ubl, M., & Siddle, R. (2024, 17 December). The rise of the AI crawler. Vercel. https://vercel.com/blog/the-rise-of-the-ai-crawler
IndexNow.org. (n.d., accessed 15 October 2026). IndexNow. https://www.indexnow.org/
SparkToro. (2026, 11 March). Office Hours: Can you actually track AI visibility? https://sparktoro.com/blog/office-hours-can-you-actually-track-ai-visibility/
Microsoft. (n.d., accessed 15 October 2026). AI Performance. Bing Webmaster Tools Help. https://www.bing.com/webmasters/help/ai-performance-9f8e7d6c
Google. (n.d., rollout note dated 31 August 2026). Generative AI performance report (Search). Search Console Help. https://support.google.com/webmasters/answer/16984139?hl=en
Google. (n.d., accessed 15 October 2026). Search generative AI control. Search Console Help. https://support.google.com/webmasters/answer/16908024?hl=en
Google. (n.d., accessed 15 October 2026). [GA4] Default channel group. Google Analytics Help. https://support.google.com/analytics/answer/9756891?hl=en
Google. (2025, 11 September). Search Quality Rater Guidelines, section 3.4. https://static.googleusercontent.com/media/guidelines.raterhub.com/en//searchqualityevaluatorguidelines.pdf
Google. (2026, last updated 5 October). Creating helpful, reliable, people-first content. Google Search Central. https://developers.google.com/search/docs/fundamentals/creating-helpful-content
Google. (2025, last updated 10 December). Introduction to structured data markup in Google Search. Google Search Central. https://developers.google.com/search/docs/appearance/structured-data/intro-structured-data
W3C. (2020, 16 July). JSON-LD 1.1 (W3C Recommendation). https://www.w3.org/TR/json-ld11/
Linehan, L. (2026, 11 May). We Tracked 1,885 Pages Adding Schema. AI Citations Barely Moved. Ahrefs. https://ahrefs.com/blog/schema-ai-citations/
GoGoChimp. (2026, 7 October). Benchmark re-crawl: 299 sites, 244 scored, scoring version 2026-10-01, up to 40 pages each, GoGoChimp's own sites excluded. First-party data. https://www.gogochimp.com/research/state-of-ai-citability-2026
Schwartz, B. (2019, 31 July). Google: There Is No Such Thing As LSI Keywords (quoting John Mueller). Search Engine Roundtable. https://www.seroundtable.com/google-lsi-keywords-27970.html
Jose, J. (2021, 19 April). More details about the page experience update. Google Search Central Blog. https://developers.google.com/search/blog/2021/04/more-details-page-experience
Dictionary.com. (n.d., accessed 15 October 2026). SEM. https://www.dictionary.com/browse/sem
Southern, M. G. (2026, October). Google's Illyes: Robots.txt Covers Crawling, Not AI Training. Search Engine Journal. https://www.searchenginejournal.com/google-illyes-robots-txt-crawling-ai-training/593957/