AI Search Statistics 2026: The State of AI Search Every Marketer Should Know

Data ages fast in AI search. The numbers that were true in early 2025 are already out of date by late 2026. What follows is a compiled, source-linked set of statistics on how big AI search has gotten, what it is doing to click-through rates, and how volatile brand visibility inside AI answers really is.

Every figure below links directly to the original study, official blog post, or company disclosure it came from. Where a number could not be traced to a primary source, it was left out rather than repeated secondhand.

How Big AI Search Has Gotten

The scale numbers moved fast through 2025 and into 2026.

• ChatGPT passed 800 million weekly active users, as stated directly by the company behind ChatGPT in its own announcement of its business milestones 1, up from 500 million in March 2025.
• Google's Gemini app surpassed 650 million monthly users, according to Google and Alphabet's own CEO writing on Google's blog 2, up from 450 million in July and 350 million in March 2025.
• Google AI Overviews had 2 billion monthly users, and Google's AI Mode had already crossed 100 million monthly active users in the US and India alone, according to Google and Alphabet's own CEO's Q2 2025 earnings remarks published on Google's blog 3.
• Google itself said AI Overviews were driving over 10% more Google Search usage for the query types that show them, a finding it published directly on its own blog 4.

Put together: hundreds of millions of people are now routing everyday questions through AI assistants instead of, or in addition to, a traditional search box. This is no longer early-adopter behavior.

What AI Answers Are Doing to Click-Through Rates

The most closely watched question in search marketing is what happens to a page's clicks once an AI Overview appears above it. Two independent research groups have now measured it directly.

Pew Research Center tracked the actual browsing activity of 900 US adults for a full month and found that 58% of them conducted at least one Google search in March 2025 that produced an AI-generated summary 6. Roughly 18% of the Google searches Pew tracked triggered an AI Overview, and users clicked a link in just 8% of searches that showed an AI summary, compared with 15% of searches that did not. Clicking through to a source cited inside the summary itself was rarer still, at just 1% of visits.

Ahrefs took a different approach, comparing the click-through rate of the top-ranking page for 300,000 keywords, with and without an AI Overview present. Its first study, published in April 2025, found AI Overviews correlated with a 34.5% lower average click-through rate for the page ranking first 7. When it reran the same methodology on December 2025 data, the gap had widened to a 58% lower average click-through rate 8.

BrightEdge, tracking AI Overview presence daily across its own keyword set, found that AI Overview coverage grew 58% year over year and now appears on 48% of tracked keywords, meaning organic-only results still show up on the other 52% 9.

The pattern across all three data sets agrees: the more a query can be answered directly, the more traffic Google now keeps for itself instead of sending it to a website.

AI Referral Traffic Is Growing, Even If It Is Still Small

Fewer clicks from Google does not mean zero traffic from AI. It means the traffic is arriving through a different route.

Similarweb measured that AI platforms sent over 1.13 billion referral visits to the top 1,000 websites globally in June 2025, up 357% year over year, in its own published analysis 10.

Retail felt this most directly. Adobe's analysis of the 2025 holiday shopping season, drawn from over a trillion site visits it tracks through Adobe Analytics, found that generative AI referral traffic to US retail sites grew 693.4% year over year across November and December 2025 12.

None of this traffic rivals what Google Search still sends. But the growth curve is steep, and it is compounding on a small base rather than shrinking.

Brand Visibility Inside AI Answers Is Volatile

Getting mentioned once inside an AI answer does not mean staying visible. AirOps analyzed roughly 45,000 citations gathered across repeated runs of the same queries and found that only 30% of brands remained visible in back-to-back AI responses to an identical question 13. Its follow-up 2026 report extended the test to five consecutive runs of the same query and found that just 20% of brands stayed present across all five 14.

The same 2026 AirOps report found that credibility for that visibility mostly does not come from a brand's own website: about 48% of citations traced back to community platforms such as Reddit and YouTube, and 85% of brand mentions originated from third-party pages rather than owned domains 14.

Pew's browsing study offers a matching picture of what AI Overviews actually cite: Wikipedia, YouTube, and Reddit were the most-cited sites, together accounting for about 15% of all links shown inside AI summaries 6.

Two things follow from this. First, appearing once in an AI answer is not the same as being reliably visible, brands flicker in and out even when nothing about the underlying content has changed. Second, third-party validation, being discussed on Reddit, reviewed on YouTube, or referenced from Wikipedia, appears to matter more to AI platforms than what a brand publishes on its own site.

Why This Matters for a 2026 Marketing Plan

A few conclusions follow directly from the numbers above, without needing to round or extrapolate them:

1. AI assistants are now mainstream, not experimental. With ChatGPT above 800 million weekly users and Gemini above 650 million monthly users, most consumers have used one of these platforms.
2. Click-through rate declines on AI-Overview-triggering queries are large and getting larger, from 34.5% (source) to 58% (source) inside a single year of Ahrefs' methodology. Traffic forecasts that assume flat organic CTR are already out of date.
3. AI referral traffic is growing fast off a small base, and retail in particular is seeing triple-digit year-over-year growth. It will not replace organic search traffic soon, but ignoring it means missing a fast-growing acquisition path.
4. Visibility inside AI answers is not a one-time achievement. With as few as 20% of brands persisting across five repeated runs of the same query (source), ongoing monitoring matters more than a single optimization pass.
5. Third-party mentions carry more weight than owned content. Getting discussed on Reddit, reviewed on YouTube, or referenced by independent sources appears to move AI visibility more than publishing on your own domain alone.

Industry forecasters have been anticipating this shift for a while. Gartner predicted in February 2024 that traditional search engine volume would drop 25% by 2026 as generative AI tools absorb queries that used to go to search engines 15. The data above suggests that shift is well underway.

For the operational side of acting on this, see our related pieces: How AI Overview Is Killing Click-Through Rates (and What to Do About It), AI Visibility Explained, and The 90-Day AI SEO Roadmap: From Zero to AI-Visible.

SKYA tracks how a brand appears in answers from AI platforms such as ChatGPT, Gemini, Perplexity and Google AI Overviews, and helps teams act on gaps in that visibility. Learn more at skya.one.

Sources

This piece compiles data directly from primary sources, including the company behind ChatGPT's own blog post disclosing its user figures, Google and Alphabet's own blog posts on Gemini's growth and Q2 2025 earnings remarks, Google's own Search blog, Pew Research Center's original browsing-panel study, Ahrefs' two independent click-through-rate studies, BrightEdge's year-over-year AI Overview tracking, Similarweb's own published referral-traffic analysis, Adobe Analytics' 2025 holiday shopping report, AirOps' citation-volatility research, and Gartner's 2024 press release forecast. Each figure is linked to its original source above.