Most AI search statistics floating around the web are stale, unsourced, or quietly made up. This page is the opposite. Every figure below carries a date and a primary source, and we refresh it as the data moves. Bookmark it. When you need a number to brief a client, build a deck, or settle an argument about where search is heading, start here.
A quick word on method, because it matters. We checked each stat against the original study or company announcement, not a secondhand blog. Where a number is contested, vendor-reported, or measured in a particular way, we say so. And we left out the popular figures that fall apart under scrutiny, like the "71% of Americans use AI for search" line that keeps circulating without a real source. As Indonesia's generative engine optimization pioneer since 2023, Arfadia reads this data daily, because our clients' visibility depends on getting it right.
How Big Is AI Search Right Now?
Start with reach, because the scale is the whole reason any of this matters. ChatGPT crossed 900 million weekly active users, confirmed by OpenAI in February 2026. Google's AI Overviews now touch roughly 1.5 billion monthly users, and AI Mode has passed a billion monthly users across more than 200 countries. This is not a niche channel anymore. It is most of the internet's search behavior, quietly shifting under everyone's feet.
The Traffic Shift Nobody Can Ignore
Here is where it gets uncomfortable for anyone who lives on organic clicks. Gartner called it back in February 2024: a 25% drop in traditional search volume by 2026, and a 50% fall in organic traffic by 2028 as people move to chatbots and AI assistants. Two years on, that read looks conservative. The mechanism behind it is the zero-click answer. Around 58.5% of US Google searches already finish without a single click, per SparkToro and Datos. Put an AI Overview on the page and that figure climbs sharply.
The squeeze on the top result is just as stark. Ahrefs found the number-one organic listing loses up to 58% of its click-through when an AI Overview sits above it, up from 34.5% a year earlier. Seer Interactive measured a similar 61% drop. And Pew Research found that with an AI summary present, people tap a citation only about 1% of the time. Fewer clicks per search, right across the board.
When AI Traffic Does Arrive, It Converts
Fewer clicks, yes. But the clicks that come through behave nothing like ordinary search traffic, and the gap is dramatic.
The reason is simple enough. Someone arriving from an AI answer has already done their research inside the chat, so they land much closer to a decision. Ahrefs put hard numbers on it: AI search visitors made up just 0.5% of traffic but drove 12.1% of signups, a ratio of roughly 24 to 1. A smaller stream, and a far hotter one.
Ranking Is Not Citation
This is the stat that breaks the most SEO assumptions. Where you rank on Google barely predicts whether AI cites you.
Only about 12% of AI-cited pages sit in Google's top 10, and roughly 80% do not rank in the top 100 at all. The number-one position earns just a 33% citation rate. One big exception, though, and it is worth holding onto. Google's own AI Overviews pull about 94% of their citations from the top 20, so for that surface specifically, classic ranking still counts a great deal.
What Actually Earns a Citation
If ranking is not the lever, what is? The research points in a remarkably consistent direction.
That controlled study from Princeton, Georgia Tech, and IIT Delhi (KDD 2024) is the cleanest evidence we have. Adding statistics lifted visibility about 41%. Citing sources and adding quotations each landed in the 30 to 41% range. Keyword stuffing did worse than changing nothing. Two more signals carry real weight. Freshness, first: Seer found recently updated content gets cited around 4.3 times more often, and roughly 85% of AI Overview citations point to content less than two years old. Then there is brand presence, which may be the biggest factor of all.
Across several 2026 analyses, branded web mentions correlated about three times more strongly with AI visibility than backlinks did, roughly 0.66 against 0.22. The takeaway writes itself. Being talked about across the web beats being linked to.
Where AI Looks for Its Sources
AI engines do not read the same corners of the web, and the citation data shows clear favorites. Earned media dominates the picture. Muck Rack found about 82% of AI citations come from earned media, and roughly 94% from non-paid sources. AirOps reported that around 85% of citations point to external sites rather than a brand's own pages. Beyond that, each engine leans a particular way.
| AI engine | Sources it leans on most |
|---|---|
| ChatGPT | Wikipedia heavily, plus established reference sites |
| Google AI Overviews | Reddit and YouTube, alongside top-ranking pages |
| Perplexity | Reddit very heavily, plus citation-rich sources |
Source preferences from 2026 citation-pattern studies. Your off-site presence should follow where each engine actually looks.
The Catch: AI Answers Are Slippery
One thing before you treat any single reading as gospel. AI answers are non-deterministic, and as sources they are honestly a little unreliable. Around 60% of ChatGPT queries get answered straight from memory, with no live retrieval at all. Citation volatility runs 40 to 60% month to month. And in one analysis, somewhere between 50 and 90% of LLM citations did not fully support the claim they were attached to. Measure the distribution over time. A snapshot will mislead you.
Visibility Is Not the Same as Trust
The newest wrinkle is the one brands should care about most. A June 2026 study from Burson and Profound, covering 85 companies and more than 55,000 believability forecasts, found that showing up in AI answers does not mean being believed. They called it the credibility paradox. Visibility and credibility are separate scoreboards. You can sit in the answer everywhere and still lose, if the way the model frames you does not earn trust. Which is exactly why sentiment and factual accuracy belong right next to reach in any serious measurement setup. For how we tie all of this back to revenue, see our AI Citation Rate Report, built on the RoGEO framework.
AI Search and GEO Statistics 2026: Quick Reference
The full set in one place, each with its source and year. Use it, cite it, and check the source before you quote it anywhere that matters.
| Statistic | Figure | Source (year) |
|---|---|---|
| ChatGPT weekly active users | 900 million | OpenAI (2026) |
| Google AI Overviews monthly reach | ~1.5 billion | Google (2026) |
| Google AI Mode users | 1 billion+ | Google (2026) |
| Searches showing an AI Overview | ~25% | Industry data (2026) |
| Forecast drop in traditional search volume | -25% by 2026 | Gartner (2024) |
| Forecast drop in organic traffic | -50% by 2028 | Gartner (2024) |
| US searches ending without a click | 58.5% | SparkToro, Datos (2024) |
| Zero-click rate when an AI Overview appears | ~83% | Similarweb (2025) |
| CTR drop for #1 result under an AI Overview | up to 58% | Ahrefs (2025) |
| Citation click rate when an AI summary shows | ~1% | Pew Research (2025) |
| AI search visitors: traffic vs signups | 0.5% vs 12.1% | Ahrefs (2026) |
| Top AI-source conversion rate (Claude) | up to 16.8% | Seer, First Page Sage (2026) |
| Google organic conversion (benchmark) | 1.76% | First Page Sage (2026) |
| AI-cited pages NOT in Google top 100 | ~80% | Ahrefs (2026) |
| Citation rate of the #1 organic result | ~33% | Ahrefs, ZipTie (2026) |
| AI Overview citations from the top 20 | ~94% | Ahrefs (2026) |
| Query fan-out share of AI citations | 51% | Search Engine Land (2025) |
| Visibility lift from adding statistics | +41% | Princeton et al., KDD (2024) |
| Citation lift from fresh content | ~4.3x | Seer (2026) |
| Brand mentions vs backlinks (AI visibility) | 0.66 vs 0.22 | Industry analysis (2026) |
| AI citations from earned media | ~82% | Muck Rack (2026) |
| ChatGPT queries answered from memory | ~60% | Industry analysis (2026) |
For the Indonesian market in particular, our own original research goes deeper. The State of SEO in Indonesia 2026 surveys local businesses on spending and AI adoption, and the Indonesia Digital Marketing Benchmark 2026 breaks down KPIs across 12 sectors.
Frequently Asked Questions
What percentage of Google searches now show an AI Overview?
Around 25% of searches surface an AI Overview as of 2026, and the share keeps growing. Prevalence varies a lot by query type, with informational and how-to searches far more likely to trigger one than navigational or transactional queries.
Is AI really reducing website traffic?
The data says yes, at least for clicks. About 58.5% of US searches already end without a click, and that rises to roughly 83% when an AI Overview is present. Gartner forecast a 25% drop in traditional search volume by 2026. The nuance is that AI-referred visitors, while fewer, convert far better than typical organic traffic.
Does ranking number one on Google get you cited by AI?
Not reliably. Across one 15,000-prompt study, only about 12% of AI-cited pages ranked in Google's top 10, and around 80% did not rank in the top 100. The #1 result earns roughly a 33% citation rate. Google's own AI Overviews are the exception, pulling about 94% of citations from the top 20.
What converts better, AI traffic or Google organic?
AI traffic, by a wide margin in the available data. Visitors from AI sources have converted at rates from about 10.5% (Perplexity) up to 16.8% (Claude), against roughly 1.76% for Google organic. Ahrefs also found AI visitors drove 12.1% of signups while making up just 0.5% of traffic.
What is the single most important factor for AI visibility?
No one factor decides it, but brand presence stands out. Branded web mentions correlated about three times more strongly with AI visibility than backlinks in 2026 analyses, and brand search volume ranked among the strongest predictors of citations. Adding original statistics, citing sources, and keeping content fresh all help too.
About Arfadia
PT Arfadia Digital Indonesia is a full-service digital marketing agency operating since 2008 and Indonesia's Generative Engine Optimization pioneer since 2023. The agency holds triple ISO certification (9001, 14001, and OHSAS 18001), partners with Google, Meta, and TikTok, and sits on the Forbes Agency Council. Our work has driven results like Toffin at plus 260 percent organic traffic with 334 AI citations in six months, Allianz at plus 380 percent, and SERA (Astra Group) at plus 1,200 percent. Explore our GEO services or read the complete GEO guide.
Sources and further reading
- OpenAI usage announcement (February 2026); Google Search announcements on AI Overviews and AI Mode (2026)
- Gartner, search volume and organic traffic forecasts (February 2024)
- SparkToro and Datos, 2024 Zero-Click Search Study; Similarweb zero-click data
- Ahrefs, AI citation, AI Overview click-through, and AI traffic studies (2025 to 2026)
- Seer Interactive and First Page Sage, AI Overview, conversion, and freshness analyses
- Pew Research, AI Overview click behavior (2025)
- Search Engine Land, query fan-out and AI citation analysis
- Aggarwal et al., Generative Engine Optimization, KDD 2024 (Princeton, Georgia Tech, IIT Delhi)
- Muck Rack, citation source analysis; AirOps, external citation data; ZipTie, ranking and citation studies
- Burson with Profound, The Credibility Paradox (June 2026); Google Search Central, AI Features and Your Website (2026)