This is a fully-sourced reference on how people use Google AI Mode, drawn from Google's official report "How people are using AI Mode in the U.S.", published at the surface's one-year mark in mid-2026. The headline: Google says AI Mode has passed 1 billion monthly active users globally, queries have more than doubled every quarter since launch, and the average AI Mode query is triple the length of a traditional search. This page collects the report's figures, each tied to its source.
Last reviewed: 22 June 2026. All figures below are from Google's AI Mode U.S. Insights report. Google states the data is from internal Google Search and Google Trends, covering the period from AI Mode's US launch in May 2025 to April 2026, and that AI Mode Trends data is not publicly available on trends.google.com. It is platform-reported and not independently audited, so read the direction as firmer than any single number, and hedge anything you quote as "as of mid-2026, per Google."
How big is AI Mode, and how fast is it growing?
Google frames AI Mode as the most significant change to Search in its history. The scale figures it reported:
| Metric | Figure | Source |
|---|---|---|
| Monthly active users (global) | Surpassed 1 billion | Google, AI Mode U.S. Insights (2026) |
| Query growth | More than doubled every quarter since launch | Google, AI Mode U.S. Insights (2026) |
| US launch | May 2025 | Google, AI Mode U.S. Insights (2026) |
The same 1-billion figure echoes what Google reported at I/O 2026 (see the AI search statistics reference). The direction is unambiguous: AI Mode now operates at the scale of the open web.
How is search behaviour changing in AI Mode?
People search differently, not just more. Google's behavioural findings:
- Queries are ~3× longer than a traditional search query — people express full intent in natural language instead of keywords.
- Follow-up queries grew more than 40% per month on average in the US, as people refine and dive deeper across turns — a live example of query fan-out playing out conversationally.
- More than 1 in 6 searches are multimodal (non-text); image-input searches grew more than 40% month-over-month since launch, making them one of the fastest-growing query types.
The lead-in words confirm the shift to question-shaped search. Google's top first words in AI Mode queries were What, How, I, Is, Can, and the top keywords were Find, Information, Identify, Explain, Summarize. People ask questions; they no longer translate them into keywords.
AI Mode rewards content that answers a full question and its follow-ups — because the average query is triple the length of a keyword search and people refine conversationally, turn after turn.
What are people doing in AI Mode?
Google groups usage into five intents — Explore, Decide, Learn, Create, Do — and reports which categories grew fastest against AI Mode queries overall:
| Behaviour | Growth (vs AI Mode queries overall) | Signal |
|---|---|---|
| Planning ("Do") | +80% faster | Logistics, schedules, to-do and exercise plans |
| "Which" decision queries | +40% faster | "Which of", "which one" grew most — comparison intent |
| Brainstorming ("Explore") | +30% faster | Open-ended, inspiration-led discovery |
| Image creation ("Create") | More than tripled since the start of the year | Creative generation inside Search |
The top 10 topics people search in AI Mode, per Google: Creative/Content Generation, Media/Entertainment, Education/Knowledge, Personal/Professional Development, Fashion/Beauty, Productivity/Practical Tasks, Food/Drinks, Technology/Coding, Health/Wellness, and Travel. The takeaway for brands: AI Mode is used for decision-making and planning, not just lookups — so content that helps a user decide and act is in the path of the fastest-growing query types.
What does the data say about shopping?
Google reports that when shopping, people often begin on traditional Search and click into AI Mode to dive deeper with follow-up questions. The top shopping topics where those follow-ups happen:
| Rank | Top shopping topics (follow-ups) | Top retail attributes asked about |
|---|---|---|
| 1 | Electronics | Price |
| 2 | Books / Movies / Music | Location |
| 3 | Apparel | Colour |
| 4 | Health / Beauty | Brand |
| 5 | Automotive | Availability |
| 6 | Home / Garden | Size |
| 7 | Grocery | Material |
| 8 | Home Improvement | Style |
| 9 | Toys / Games | Type |
| 10 | Sports / Outdoors | Quality |
For apparel specifically, the top follow-up subjects were Clothing, Shoes, Jewelry, Handbags, and Wedding/Bridal. The practical read: AI Mode shoppers ask for concrete, structured attributes — price, availability, size, material — so brands that expose those facts cleanly win the follow-up. We turn this into an action plan in how to optimise your products for AI Mode shopping.
How should you use these figures?
Treat this as a dated reference, not a guarantee. Every number here is Google's own, drawn from internal Search and Trends data that Google notes is sampled and not publicly verifiable on trends.google.com — so cite Google and the date when you reuse one, and hedge as "as of mid-2026." Because live retrieval favours recently-updated pages, we keep references like this on a refresh cadence and update figures substantively as new data lands.
The figure that ultimately matters is your own: whether AI Mode and the other engines surface, cite, and recommend your brand for the questions your customers actually ask — tracked over time, not spot-checked. That measurement is exactly what Buffy Intel is built to provide.
