AI and search glossary

Agentic AI is the umbrella term for systems where language models plan steps and take actions through tools, with less step-by-step direction from people.

Also called agentic systems.

An AI agent is a language model that chooses and calls tools in a loop, reading each result, until it finishes a task or hands off to a person.

Also called LLM agent or autonomous agent.

An AI copilot is an assistant built into software your team already uses that drafts and suggests work, while a person makes every final decision.

Also called embedded AI assistant.

AI guardrails are checks placed around a language model that block, rewrite or escalate inputs and outputs that break your rules on safety, accuracy, privacy or scope.

Also called LLM guardrails.

An AI hallucination is output from a language model that is fluent and confident but false, such as an invented citation, policy, figure or product detail.

Also called confabulation.

AI visibility measures how often AI assistants mention or cite your business when answering the questions your buyers ask, compared with competitors over the same set of prompts.

Also called AI search visibility or LLM visibility.

Answer engine optimization structures pages so a search engine or AI assistant can lift a direct answer from them, as a featured snippet, voice answer or AI citation.

Chunking is splitting documents into smaller passages before they are embedded and indexed, so a retrieval system can find and return the exact part that answers a question.

Also called text splitting.

A context window is the maximum amount of text, measured in tokens, that a language model can read and write in one request, counting instructions, documents and its reply.

Also called context length.

Embeddings are lists of numbers that capture the meaning of text or images, so software can measure how closely related one item is to another.

Also called vector embeddings.

Entity SEO is making sure search engines and AI systems recognize your business as one distinct thing, with consistent facts about what it is, does and connects to.

Also called entity optimization.

Fine-tuning is further training of an existing language model on your own examples, so it learns a specific format or task without long instructions every time.

Function calling lets a language model ask your software to run a named function, such as a CRM lookup, by returning the function name and arguments as JSON.

Also called tool calling.

Generative engine optimization is the work of making your pages likely to be retrieved, quoted and cited in AI-generated answers from ChatGPT, Perplexity, Gemini and Google AI Overviews.

Also called LLM optimization (LLMO).

AI Overviews are AI-generated summaries shown at the top of some Google results with links to sources; AI Mode is a conversational search view built on the same systems.

Also called Search Generative Experience (SGE).

Human-in-the-loop is a workflow design where a person reviews, corrects or approves an AI system's output at set points before it reaches a customer or a system of record.

Also called human review.

Hybrid search runs keyword search and semantic search on the same query and merges the results, so it catches exact terms and paraphrased meaning together.

Also called hybrid retrieval.

Intelligent document processing uses OCR and AI models to read documents such as invoices, contracts and forms, extract the fields you need, check them and post them into your systems.

Also called document AI.

A large language model is a neural network trained on huge volumes of text to predict the next token, which lets it write, summarize, classify and answer questions.

LLM evaluation scores a language model system's outputs against a fixed set of real test cases, so you can tell whether a change made it better or worse.

Also called LLM testing.

LLMOps is the operating discipline for language model systems in production: tracing every call, tracking cost and latency, monitoring quality, and managing prompt and model changes.

Also called LLM observability.

llms.txt is a proposed Markdown file at a site's root that gives AI tools a short site summary and a hand-picked list of links to its key pages.

The Model Context Protocol is an open standard for connecting AI applications to outside tools and data, so a single integration works across compatible assistants and agents.

n8n is a workflow automation platform, self-hosted or cloud, that connects apps and APIs in visual workflows and supports custom code and AI agent steps.

Prompt engineering is writing and testing the instructions, examples and context given to a language model so it produces the output a task needs, consistently.

Also called prompt design.

Prompt injection is an attack where text given to a language model, typed by a user or hidden in a document, overrides the instructions the system was built with.

Retrieval-augmented generation is a method where a system first searches your own documents for relevant passages, then gives them to a language model to answer from.

Also called grounded generation.

Schema markup is structured data, usually JSON-LD, that labels the facts on a page, such as organization, service, price or author, using the shared schema.org vocabulary.

Also called schema.org markup.

Semantic search finds results by meaning instead of exact words, so a query can match relevant content even when they share no vocabulary.

Also called vector search or neural search.

A vector database stores embeddings, the numeric fingerprints of text or images, and quickly finds the stored items whose meaning is closest to a query.

Also called vector store.