Google AI Mode is a conversational search mode inside Google Search.
It answers a question with one written, cited response in place of a page of links. It runs on a custom Gemini model and handles long questions with several parts, plus follow-ups. Anyone asking what is AI Mode has seen it as a tab next to "All" on the results page.
Google AI Mode is a separate search experience in which Gemini writes the answer and the web supplies the sources. You open it from its own tab and type or speak a question. You then keep asking follow-ups in the same thread. Links to the pages it drew from sit beside the answer. Google set out its main features in the Google I/O 2025 announcement, and it has kept changing the model underneath since then.
AI Mode splits your question into smaller searches and runs them in parallel. Google calls this query fan-out. In its I/O post, Google said AI Mode works by "breaking down your question into subtopics and issuing a multitude of queries simultaneously on your behalf." The model then reads the passages it found and writes one reply that cites the pages it used. Deep Search is a heavier version of the same process, and Google says it runs hundreds of searches for a single report.
AI Mode is a full conversation you choose to open, while AI Overviews are short summaries that Google places at the top of a normal results page. Google Search Central guidance says both features use query fan-out, but each one fits a different moment in a search.
AI Overviews show up without being asked for, above the regular links, when Google's systems judge that a summary helps. The classic results are still there underneath. AI Mode replaces that page. You go into it on purpose, and the answer fills the screen.
An Overview covers one query in a few paragraphs, and that's the end of it. AI Mode remembers the context, so your second question builds on the first. It suits comparisons and planning questions that otherwise require a string of separate searches.
AI Mode moves the click away from a ranked link and toward a source cited inside an answer. Readers who get what they need from the reply skip a site visit, so impressions rise while clicks stay flat. Sub-queries also reach pages that don't rank for the main keyword, because the system judges individual passages. Google counts these visits in the Performance report's "Web" type together with regular results, so read the trends there with that in mind.
To be cited in AI Mode, a page has to be indexed and eligible to appear in Google Search with a snippet. Google's own documentation says there are "no additional technical requirements" beyond that. Your page gets picked when one of its passages answers a sub-question the model searched for, and answers it clearly enough to quote. If you block snippets with a nosnippet rule, the page drops out of the supporting links.
Optimising for AI Mode means you answer the questions a searcher asks next as well as the main one. In practice:
To check your work, ask AI Mode your target questions and note which pages it cites. Then compare those pages with yours.
These terms come up next to AI Mode, and each one means something narrower or broader:
Google updates the model behind AI Mode regularly. If you need the current answer to what is AI Mode, start with Google's Search Central documentation, then test your own questions in the tab.