Back to Glossary

AI & Modern Search

AI Overviews, semantic search, LLM visibility, knowledge graphs, and retrieval-driven search.

Start the SEO path Browse all terms

Reference library

All Terms

11 terms

A

AI Content DetectionAI content detection describes tools and methods that estimate whether text was created with AI. For SEO, the more important question is whether the content is helpful, accurate, original, and responsibly reviewed.Intermediate AI HallucinationAI hallucination describes AI output that sounds plausible but is false, unsupported, fabricated, or misleading. In SEO, the main risks are wrong facts, invented sources, and confident claims without reliable evidence.Advanced AI OverviewsAI Overviews are AI-generated summaries in Google Search that answer selected queries with a short synthesis and links for deeper exploration. For SEO, the implication is simple: a page should not only rank, but explain clearly enough to be useful as a cited source.Intermediate

C

ChatGPT SearchChatGPT Search is the web search capability inside ChatGPT. It retrieves current information from the web, writes an answer with cited sources, and links to the pages it used.Intermediate

K

Knowledge GraphA knowledge graph is a knowledge structure of entities, properties, and relationships that helps search systems understand real things instead of only words.Intermediate

L

LLM VisibilityLLM visibility describes how visible a brand, website, source, or product category is inside large language model answers.Advanced

P

Passage RankingPassage ranking helps Google understand individual sections of a page and evaluate their relevance for specific searches.Intermediate

Q

Query Fan-OutQuery fan-out is a retrieval technique in which an AI search system splits one question into several related sub-queries, runs them across subtopics and sources, and combines the results into one answer.Advanced

R

Retrieval-Augmented Generation (RAG)Retrieval-Augmented Generation, or RAG, is an AI pattern where a language model retrieves relevant sources before generating an answer. The output is therefore grounded in external context such as documents, databases, or knowledge bases, not only in model memory.Advanced

S

Semantic SearchSemantic search means search systems try to understand meaning, context, entities, and intent instead of matching words only. The question is no longer just whether the keyword appears. It is whether the page answers the topic in the right meaning.Advanced

V

Vector SearchVector search finds content through mathematical closeness in a meaning space rather than exact word matches. Text is turned into embeddings. Similar meanings are close together and can be retrieved even when the words differ.Advanced