AEO now has more vocabulary than SEO did in 2010. New acronyms show up every quarter, some old ones quietly redefined, and vendor-specific terms multiply across ChatGPT, Perplexity, Claude, Gemini, and Copilot. If you sit in a category strategy meeting where AEO, GEO, LLMO, MCP, and RAG all get thrown into the same sentence, this reference is for you.
Sixty-five terms, grouped into seven categories, one or two lines each. Bookmark it, share it with product and marketing, and quit re-Googling every acronym. Everything below is current as of July 2026.
AEO Fundamentals
1. AEO (Answer Engine Optimization). Structuring content so AI-powered answer systems select it as a cited source. The umbrella discipline covering generative and non-generative answer surfaces.
2. GEO (Generative Engine Optimization). The subset of AEO focused specifically on earning citations inside synthesized AI answers from ChatGPT, Perplexity, and Google AI Overviews.
3. LLMO (Large Language Model Optimization). The widest of the three overlapping acronyms, covering any strategy aimed at visibility inside LLM-based systems, including agents and enterprise chatbots.
4. AIO (AI Optimization). Informal catch-all used interchangeably with AEO or GEO, sometimes narrowed to strategies aimed specifically at Google AI Overviews inclusion.
5. AI search. Information retrieval where a language model synthesizes a direct answer rather than returning ranked links. The query is treated as a prompt.
6. Answer engine. Any system designed to return a single synthesized response to a natural-language query. Examples include Perplexity, ChatGPT Search, and Google AI Mode.
7. Citation. An explicit reference an AI system makes to a source document when producing a response. May appear as a footnote, inline link, or source card.
8. Retrieval. The step in an AI search pipeline where the system fetches relevant documents before the language model generates its answer. Retrieval quality shapes which sources can appear in output.
9. RAG (Retrieval-Augmented Generation). An AI architecture that connects a language model to an external knowledge source at inference time, grounding output in retrieved documents rather than trained parameters.
10. LLM (Large Language Model). A neural network trained on large-scale text data to predict and generate language. GPT-4o, Claude, and Gemini are current examples.
11. Foundation model. A large-scale pre-trained AI system intended as a base that developers fine-tune for specific applications. All LLMs are foundation models; the category also includes image, audio, and video models.
12. Prompt. The natural-language input a user submits to an answer engine, serving the function that a keyword query serves in traditional search.
AI Search Engines and Interfaces
13. ChatGPT Search. OpenAI's real-time web search feature inside ChatGPT, using the OAI-SearchBot crawler at query time. Available to Plus, Team, and Enterprise subscribers.
14. Perplexity. A standalone answer engine that retrieves live web results and presents synthesized responses with numbered source cards. One of the highest-volume AI referrers to publisher sites.
15. Claude with Search. Anthropic's web-retrieval mode for Claude, connecting to current sources via the Claude-SearchBot crawler before generating a grounded answer.
16. Gemini. Google's family of multimodal foundation models, powering Google AI Overviews, Google AI Mode, and the standalone Gemini assistant.
17. Copilot. Microsoft's AI assistant integrated into Bing, Microsoft 365, and Windows. Built on OpenAI models with Bing's index as the retrieval layer.
18. Google AI Mode. A separate chat-based search interface within Google Search for complex queries. AI Mode and AI Overviews cite the same URLs only 13.7% of the time.
19. Google AI Overviews. AI-generated answer panels above standard Google results, evolved from the Search Generative Experience (SGE) experiment. Reached 2.5 billion monthly active users by May 2026.
20. Grounded response. An AI-generated answer where every factual claim can be traced back to a specific retrieved source. Measured as the proportion of statements supported by retrieved context.
21. Agentic search. A retrieval pattern where an AI agent autonomously plans a multi-step research process, calling tools and fetching sources iteratively until it has enough evidence.
22. Zero-click search. Any search session that ends on the results page without the user clicking through to a source website. AI answers accelerate zero-click behavior.
200+ SaaS teams already track their AI citations.
They know exactly when ChatGPT mentions their brand, and when it stops. Do you?
Crawlers and Indexing
23. GPTBot. OpenAI's training data crawler. Distinct from OAI-SearchBot, which handles ChatGPT Search retrieval. Blocking GPTBot affects training only.
24. ClaudeBot. Anthropic's training data crawler. Claude-SearchBot handles live Claude with Search retrieval; the two are separate user-agent strings.
25. PerplexityBot. Perplexity's crawler, used for both indexing and real-time fetching. Blocking it removes a site from Perplexity's citation pool entirely.
26. Google-Extended. A robots.txt opt-out token letting site owners exclude content from Gemini training without affecting Google Search. Has no dedicated crawler behind it.
27. Applebot-Extended. Apple's robots.txt control token for excluding content from Apple Intelligence training while keeping the site accessible to Applebot for standard indexing.
28. robots.txt. The plain-text file at a domain root that communicates crawl permissions to web robots using the Robots Exclusion Protocol. See the full directive reference for syntax.
29. llms.txt. A community-proposed convention from Jeremy Howard (Answer.AI) where a site places a Markdown index of key resources at its root. Not a formal standard.
30. ai.txt. A proposed extension of robots.txt semantics for AI crawler permissions, giving finer-grained control by purpose (training vs retrieval). Still a draft, minimal adoption.
31. Crawl budget. The number of URLs a crawler will fetch from a domain in a given window. Pages that exceed budget may not enter AI citation pools in time.
32. Index. In AI search, the data store an answer engine searches during retrieval, often containing vector embeddings alongside metadata rather than only keyword-inverted lists.
Content Structure for AI
33. Schema markup. Structured data annotating content with Schema.org vocabulary. Pages with schema are cited two to three times more often by AI engines than equivalent unstructured pages.
34. JSON-LD. The structured data serialization format Google recommends for embedding Schema.org annotations. Placed inside a <script type="application/ld+json"> tag.
35. FAQ schema. A Schema.org Question/Answer markup type. The single highest-impact schema type for AI citation in 2026 because it aligns page structure with the query format.
36. HowTo schema. A Schema.org type that marks up step-by-step instructional content, tagging each step with name, description, and image so AI engines extract the process cleanly.
37. DefinedTerm schema. A Schema.org type for glossary entries and definitions. AI systems treat DefinedTerm markup as a strong signal that a page is a reliable definitional source.
38. Entity. A distinct, identifiable thing (person, organization, product, concept) with properties and relationships representable in structured form. Pages that describe a known entity get retrieved more often.
39. Ontology. A machine-readable specification of concepts within a domain and the relationships between them. Schema.org is a shared simplified ontology; enterprise deployments layer domain-specific ones on top.
40. Structured data. Any content formatted so machines can parse its meaning without inference. Acts as an explicit contract between publisher and retrieval system about what a page contains.
41. Named-entity recognition (NER). An NLP technique that automatically identifies and categorizes specific entities within unstructured text. Retrievers use NER to link page content to knowledge-graph entities.
42. Semantic HTML. The use of HTML5 elements like <article>, <section>, <header>, <figure> that carry meaning about content role. AI crawlers use it to locate the main body without full visual rendering.
How often does ChatGPT mention your brand?
Most founders have no idea. The answer might surprise you.
Citation Metrics
43. Share of voice (SOV). Your brand's citation count as a percentage of all brand citations across a fixed prompt set on a given engine. Calculated as (your citations / total category citations) x 100.
44. Citation rate. The proportion of tested queries where an AI engine cites your domain at least once. Strong B2B benchmarks target 10-15% on category queries; leaders exceed 30%.
45. Brand mention frequency. The average number of times a brand name appears in AI responses across a prompt set. Captures ambient presence even without a hyperlinked citation.
46. Source authority. An AI engine's internal assessment of how credible a domain is as a citation candidate. Informed by backlinks, EEAT signals, structured data, and historical citation patterns.
47. Retrieval score. The similarity value an AI system assigns to a candidate document during retrieval, typically cosine similarity between query and document embeddings. Higher scores enter the answer context window.
48. Citation graph. A directed network where nodes are documents and edges are citation relationships. Used to identify authoritative hubs and measure citation flow across a content ecosystem.
49. Provenance. The documented chain of sources an AI drew on to produce a specific claim. Poor provenance tracking is a primary cause of attribution hallucinations.
50. Attribution. Linking a claim in an AI answer back to the source document it came from. Requires both correct retrieval and correct passage-to-claim mapping.
51. Hallucination. An AI output that sounds plausible but is factually wrong, unsupported by retrieved sources, or wrongly attributed. Researchers distinguish factual, grounding, citation, and reasoning subtypes.
52. Groundedness. The degree to which an AI answer is supported by retrieved documents rather than inferred from model weights. A fully grounded response maps every claim to a specific retrieved passage.
Business and Buyer Terms
53. B2B AEO. Applying answer engine optimization to business-to-business sales contexts, where buyers use AI assistants to research vendors and build shortlists before contacting sales.
54. Enterprise LLM adoption. Organizational integration of LLM tools into workflows, from internal knowledge management to procurement research. As of 2026, 94% of B2B buyers report using LLMs for vendor research.
55. MCP (Model Context Protocol). An open standard developed by Anthropic and donated to the Linux Foundation's Agentic AI Foundation in December 2025, defining how AI agents connect to external tools and data sources at runtime.
56. Agentic buyer. A business purchaser who delegates early-stage vendor research, shortlisting, or RFP drafting to an AI agent rather than doing it manually.
57. AI-first buyer journey. A purchasing process that begins with an AI assistant prompt rather than a search query or analyst report. Time-to-consideration is 40-60% shorter than traditional cycles.
58. LLM procurement. Use of LLM assistants inside a formal purchasing process, from needs analysis and vendor discovery through shortlisting and price comparison.
59. Zero-click marketing. Structuring brand content so it delivers value inside AI answers even when users never visit your site. Success metric shifts from referral traffic to citation rate.
60. GEO spend. The portion of a marketing budget allocated to generative engine optimization, including content creation, schema implementation, authority outreach, and AI citation monitoring. Distinct from traditional SEO spend.
Related SEO Concepts
61. EEAT (Experience, Expertise, Authoritativeness, Trust). Google's four-part quality framework from the Search Quality Evaluator Guidelines. AI engines use EEAT-adjacent signals to rank citation candidates.
62. Helpful Content. Google's content quality standard from 2022 that rewards content written for people rather than search engines. AI engines apply a similar filter on direct-answer relevance.
63. Semantic search. A retrieval approach that matches queries to documents based on conceptual meaning rather than exact keyword overlap. The foundation all major AI answer engines build on.
64. Embedding. A dense numerical vector representation of text, generated so semantically similar content sits close together in high-dimensional space. Both query and documents get embedded for retrieval.
65. Vector database. A data store optimized for embeddings, enabling fast approximate nearest-neighbor searches across millions of documents. The retrieval backend most AI search engines run on.
15 hours a week manually. Or 15 minutes with RankControl.
Track citations, monitor competitors, and fix content gaps across every AI search engine. Automatically.
The vocabulary keeps expanding because the surface area does. Next quarter will add three more crawler names, two more measurement metrics, and at least one new acronym for the same idea somebody already coined. Keep this sheet handy, drop the terms into the right slots, and the next AEO meeting your team runs will finish in half the time. If you want the vendor-by-vendor bot list that pairs with the crawler section, we published the complete AI crawler list as a separate reference.
