Google’s AI search is becoming the default experience. Similarweb says AI Overviews now appear on more than 40% of U.S. searches in 2026, while Google says AI Mode has surpassed 1 billion monthly active users globally.
The product shift is not just that Google added more AI summaries. Google now lets users move directly from AI Overviews into AI Mode, and Similarweb describes the two as part of one seamless search flow. That makes conversational AI a built-in path inside Search, not a side feature.
Similarweb’s figures are market-intelligence estimates rather than Google-disclosed product metrics. But taken with Google’s own announcements around AI Mode, the direction is hard to miss: Google Search is being rebuilt around AI answers first, links second.
Similarweb’s 2026 data shows AI Overviews now trigger on more than 40% of U.S. searches
The clearest independent signal is Similarweb’s 2026 Generative AI Landscape Report, which says AI Overviews trigger on more than 40% of U.S. searches. That is a large jump from the earlier stage of rollout, when a 2026 measurement study found AI Overviews on 13.7% of trending queries overall and 64.7% of question-form queries.
Those two figures are not directly identical measurements. The arXiv study sampled trending queries and broke out question-form searches, while Similarweb’s report presents a broader U.S. search prevalence estimate. Still, the range matters: AI Overviews have moved from a query-type-heavy feature to something that now shows up across a very large share of Search.
Google’s own usage framing points the same way. In a July 2026 post, Google said the average AI Mode query is triple the length of a traditional Search query. That is a small product detail with a big implication: users are increasingly treating Search less like a keyword box and more like a chat interface.
“AI Mode has surpassed 1 billion monthly active users globally.”, Google Search’s U.S. insights post, July 2026
Google has not published comparable external methodology for that 1 billion figure. The number is still useful because it shows what Google itself considers the scale of AI-native search behavior.
Google is merging AI Overviews and AI Mode into a single conversational search flow
The product design change is straightforward. Google’s I/O 2026 Search update said AI Mode is globally default-powered by Gemini 3.5 Flash, and another Google update said users can follow up directly from AI Overviews into AI Mode, with Gemini 3 becoming the default model for AI Overviews.
In plain terms, Google has removed a lot of the friction between “classic search result with an AI summary” and “full conversational search session.” AI Overviews now work like the on-ramp, and AI Mode is the deeper lane.
That matters because user behavior tends to follow interface defaults. If the first page already answers the query in synthesized form, and the next click keeps the user inside a multi-turn AI conversation, the open web is no longer the obvious next step. That also raises practical questions about what Google stores and uses inside these interactions, including the company’s Google Search and Gemini activity settings.
Similarweb’s July 2026 update says this merger is visible in traffic patterns as well as product UI. It ties the rise of AI Overviews, longer queries, and AI Mode usage into one behavior shift rather than three separate features.
Here is the shortest version of the comparison:
| Signal | What it shows |
|---|---|
| AI Overviews on 40%+ of U.S. searches | AI answers are appearing at mass-market scale |
| AI Mode at 1 billion+ monthly active users globally | Conversational search is no longer niche |
| Average AI Mode query is 3× traditional Search length | Users are shifting from keywords to fuller prompts |
| Direct follow-up from AI Overviews into AI Mode | Google has joined summary search and chat search into one flow |
AI Overviews now work like the on-ramp, and AI Mode is the deeper lane.
Independent studies link AI search to fewer outbound clicks and weaker publisher power
The web-economy consequence is not subtle. AI search reduces the need to click out. That is the promise for users and the problem for publishers.
A 2026 arXiv study measuring Google AI Overviews looked at activation rates, source quality, claim fidelity, and publisher impact. It found AI Overviews on 13.7% of trending queries overall and 64.7% of question-form queries, which already suggested that Google was most willing to synthesize answers for the kinds of searches that once sent high-intent traffic to information sites. The paper also examined how often cited sources aligned with claims, putting pressure not just on click volume but on citation quality.
A second 2026 study on Google and Reddit found that AI Overviews sometimes increased Reddit engagement, but AI Mode largely eliminated those gains. That result is worth pausing on. AI summaries may still leak some attention to cited destinations; full conversational search appears better at keeping the session inside Google.
The academic evidence is recent, and some of it is still in preprint form rather than peer-reviewed journal publication. Even so, it matches the broader commercial pattern behind publisher defenses against AI crawling and new barriers to AI web scraping: if AI systems answer first and attribute later, the bargaining power of the underlying content supplier weakens.
Bloomberg reported the strategic backdrop in 2025. In that report, Google had decided against offering publishers separate options to appear in Search but opt out of AI search uses. For publishers, that means participation in Google Search and participation in Google’s AI layer were effectively tied together.
That linkage changes the negotiation. A publisher can still block or challenge AI access in some places, but doing so against Google has been harder because Search visibility remains economically central. The more Search becomes an answer engine instead of a referral engine, the less power publishers get from being indexed at all.
Google’s next big milestone is likely not another standalone AI feature. It is the further removal of the line between results page, summary, and chat session. Google’s I/O 2026 Search roadmap points in that direction already.
Key Takeaways
- Similarweb says AI Overviews now appear on more than 40% of U.S. searches in 2026, which makes AI-generated answers a mass-market part of Google Search.
- Google says AI Mode has surpassed 1 billion monthly active users globally, though it has not published comparable outside methodology for that figure.
- Google now lets users move directly from AI Overviews into AI Mode, turning summary search and chat search into one product flow.
- A 2026 measurement study and a separate Google-Reddit study both point toward weaker outbound traffic dynamics as AI search becomes more capable.
- Bloomberg reported in 2025 that Google chose not to give publishers a separate way to stay in Search while opting out of AI search uses.
Further Reading
- The 2026 Generative AI Landscape Report, Similarweb landing page with headline metrics on AI Overview prevalence and AI-shaped search behavior.
- AI Search News Update: July 2026, Similarweb’s summary of how AI Overviews and AI Mode are converging inside Google Search.
- Google Search’s I/O 2026 updates: AI agents and more, Google’s announcement of AI Mode’s global rollout and Search product updates.
- AI Mode in Google Search and AI Overviews get Gemini upgrades, Google’s explanation of direct follow-up from AI Overviews into AI Mode.
- How AI Mode is changing the way people search in the U.S., Google’s data-science post on AI Mode usage and query length.
- Measuring Google AI Overviews: Activation, Source Quality, Claim Fidelity, and Publisher Impact, A 2026 measurement study on where AI Overviews appear and how they cite sources.
- The Impact of AI Search on the Online Content Ecosystem: Evidence from Google and Reddit, A 2026 study on AI search effects on traffic and engagement.
- Google Decided Against Offering Publishers Options in AI Search, Bloomberg’s report on Google’s publisher policy stance.
Last reviewed: 2026-07