AI Mode
Google's conversational search experience for complex, multi-part questions, returning a reasoned AI response with supporting links…
GEO AIQuery decomposition
Query fan-out
The technique in AI search of breaking one complex user question into many related sub-queries, running them in parallel, and synthesising the results.
In full
Google has described AI Mode as issuing multiple related searches across subtopics and data sources, then assembling an answer from what they return. The consequence for site owners is that visibility depends on covering the constituent sub-questions, not just the phrasing the user typed — a page that answers one facet well can be pulled into an answer for a question it does not literally match. It also explains why AI answer citations often come from pages that do not rank for the original query.
Example
'Which of these two cities is better for a family relocation' fans out into schools, housing costs, commute times and crime, each retrieving different sources.
Related terms
Grounding
Constraining a language model's output to information retrieved from a specified source — a search index, a document set or a database —…
Retrieval-augmented generation
An architecture where a system retrieves relevant documents or passages first and then has a language model generate an answer conditioned…
Chunk retrieval
Retrieving individual segments of a document rather than whole pages, which is how most AI search and RAG systems select what to feed a…
Semantic search
Retrieval based on the meaning of a query and documents rather than literal keyword matching, using entities, context and learned…