AI search is becoming a large-scale layer of the search experience. Google reported more than 2.5 billion monthly active users for AI Overviews and more than 1 billion monthly users for AI Mode by August 31, 2026. At the same time, Pew Research Center found that AI summaries appeared in 18% of the Google searches it tracked in March 2025, and that their presence reduced clicks on traditional results.
Table of contents
- AI search adoption and global reach
- How AI is changing search queries
- AI summaries, citations and visibility
- Clicks, sessions and website traffic
- User attitudes, usefulness and trust
- AI referrals and commercial outcomes
AI search adoption and global reach
Google’s reported user figures show how quickly AI features have moved from limited experiments into mainstream search products. Google said AI Overviews had more than 1.5 billion monthly users in April 2025. By July 22, 2025, Google reported more than 2 billion monthly users across more than 200 countries and territories and 40 languages. These are Google-reported platform figures, not an independent estimate of unique people across all AI search services. (Alphabet Q1 2025 earnings call; Q2 2026 earnings call)
By August 31, 2026, Google said AI Overviews had reached more than 2.5 billion monthly active users worldwide. Google also said AI Mode had surpassed 1 billion monthly users globally. A separate Google update said AI Mode passed 1 billion monthly active users one year after its U.S. launch, measured in May 2026. (New opportunities, control and insights for website owners; How AI Mode is changing and expanding the way people search in the U.S.)
AI-assisted discovery is not limited to typed text. Google said AI Overviews were available in more than 200 countries and territories and over 40 languages on May 20, 2025. Google Lens was being used for nearly 20 billion visual searches per month in October 2024. Circle to Search was available on more than 150 million Android devices in October 2024 and on more than 250 million devices in the first quarter of 2025, when usage increased nearly 40% during the quarter. (AI Overviews expand to over 200 countries and territories, more than 40 languages; Google updates: AI-Organized Search, Google Lens, and more; Alphabet Q1 2025 earnings call)
How AI is changing search queries
Google reported that AI Overviews increased Google usage by more than 10% for the types of queries that display them in the United States and India in May 2025. That measure describes a change in usage for qualifying query types; it does not mean that every Google search increased by more than 10%. (AI Overviews expand to over 200 countries and territories, more than 40 languages)
Google’s May 2026 analysis of U.S. AI Mode also showed faster growth in more exploratory types of searches. During the six months ending in May 2026, planning queries grew 80% faster than AI Mode queries overall. Since AI Mode launched, brainstorming queries grew 30% faster than queries overall. Image searches in U.S. AI Mode grew more than 40% month over month in May 2026. These are relative growth comparisons, not counts of searches. (How AI Mode is changing and expanding the way people search in the U.S.)
Pew Research Center’s March 2025 search sample shows that query length and wording were associated with the appearance of an AI summary. Only 8% of one- or two-word Google queries produced an AI summary, compared with 53% of queries containing at least 10 words. Queries beginning with question words such as “who,” “what,” “when” or “why” produced an AI summary 60% of the time. Searches containing both a noun and a verb produced one 36% of the time. (Google users are less likely to click on links when an AI summary appears in the results)
AI summaries, citations and visibility
Pew examined 68,879 unique Google searches from March 2025. An AI summary appeared in 12,593 searches, or 18% of the sample. The result is a measured share of the searches in that study and should not be treated as a universal rate for every country, device, query category or later period.
The summaries varied substantially in length. The median summary was 67 words, while the shortest was 7 words and the longest was 369 words. Source diversity was common: 88% of the AI summaries in the sample cited at least three sources, while only 1% cited a single source. (Google users are less likely to click on links when an AI summary appears in the results)
The source mix also differed across types of search results. Wikipedia, YouTube and Reddit together accounted for 15% of the sources listed in the AI summaries Pew examined. Government websites represented 6% of sources linked in AI summaries, compared with 2% of standard search-result sources. News websites represented 5% of sources linked in both AI summaries and standard search results.
| Measure from Pew’s March 2025 sample | Result |
|---|---|
| Unique Google searches examined | 68,879 |
| Searches producing an AI summary | 12,593 |
| Share producing an AI summary | 18% |
| AI summaries citing at least three sources | 88% |
| AI summaries citing one source | 1% |
| Median summary length | 67 words |
These figures make visibility more complicated than a conventional ranking position. A page may be one of several cited sources, while the searcher receives an answer before deciding whether to open any result. The Pew figures measure citations and behavior in a defined sample; they do not establish that every cited page receives a click.
Clicks, sessions and website traffic
When a Google AI summary appeared in Pew’s March 2025 sample, users clicked a traditional Google result in 8% of visits. When no AI summary appeared, users clicked a traditional result in 15% of visits. Users clicked a link inside an AI summary in only 1% of visits with an AI summary.
The same study found that browsing sessions ended after 26% of Google pages with an AI summary, compared with 16% of pages containing only traditional results. This suggests that an AI summary can satisfy or redirect a visit without a traditional-result click, but the statistic does not identify the user’s reason for ending the session.
For publishers and SEO teams, the practical distinction is between being visible in an answer and receiving a session. A citation can support brand exposure or perceived authority even when it does not produce an immediate visit. Conversely, a lower click rate makes it more important to understand which pages, topics and query types attract follow-up visits. (Google users are less likely to click on links when an AI summary appears in the results)
User attitudes, usefulness and trust
Pew Research Center surveyed U.S. adults from August 18–24, 2025. In that survey, 65% said they encountered AI summaries in search results at least sometimes, including 45% who said they encountered them extremely often or often. Seventeen percent said they rarely or never encountered them.
Among U.S. adults who had seen AI summaries, 20% found them extremely or very useful, while 52% found them somewhat useful. The remaining 28% found them not too useful or not at all useful. On trust, 53% expressed at least some trust in the information, but only 6% said they trusted it a lot. Another 46% said they had not much trust or no trust at all.
| View among U.S. adults who had seen AI summaries | Share |
|---|---|
| Extremely or very useful | 20% |
| Somewhat useful | 52% |
| Not too or not at all useful | 28% |
| At least some trust | 53% |
| A lot of trust | 6% |
| Not much or no trust | 46% |
Age differences were pronounced in the same survey. Among U.S. adults under 30, 62% said they encountered AI summaries at least often, compared with 23% of adults 65 and older. These findings describe U.S. adults’ reported experiences during the survey period, not global attitudes or a permanent level of trust. (Americans have mixed feelings about AI summaries in search results)
AI referrals and commercial outcomes
Adobe’s Quarterly AI Traffic Report measured strong year-over-year growth in AI referrals to several types of sites during the 2025 holiday season, from November 1 through December 31, 2025. Referrals to retail sites rose 693% year over year. Travel sites rose 539%, financial-services sites rose 266%, banking sites rose 344%, and technology and software sites rose 120%.
| Site category | Year-over-year AI referral change, 2025 holiday season |
|---|---|
| Retail | 693% |
| Travel | 539% |
| Banking | 344% |
| Financial services | 266% |
| Technology and software | 120% |
Adobe also reported that AI-referred retail visits converted 31% more often than visits from non-AI sources during the holiday season. AI-referred retail visits had 45% longer time per visit, users viewed 13% more pages per visit, and the visits had a 33% lower bounce rate than non-AI traffic. Adobe measured AI-driven retail revenue per visit as 254% higher year to date and 32% higher than non-AI traffic in its 2025 holiday report.
The report included a November 2025 consumer survey as well. Among consumers using AI for online shopping, 65% said they felt more confident in their purchase and 68% said they were less likely to return the product. The referral and engagement results are Adobe measurements for the stated periods; they should not be generalized to every AI service, industry or website. (Quarterly AI Traffic Report)