AI Traffic Share Report 2026: How Much Traffic AI Search Really Captures

• Amol Ghemud
• AI Traffic Share Report 2026: How Much Traffic AI Search Really Captures
• Published: March 16, 2026

Contributors: Amol Ghemud
Published: March 16, 2026

Summary

AI chatbots sent 0.25% of the search, social and AI chatbot referrals Statcounter tracked worldwide in August 2026, up from 0.23% a year earlier and below a 0.43% peak in April and May 2026. ChatGPT sent 79.4% of AI chatbot referrals, followed by Gemini (10.9%), Perplexity (4.3%), Copilot (2.8%) and Claude (2.6%). This AI traffic share report pairs that referral data with verified zero-click studies from Pew Research Center, Ahrefs, SparkToro and Semrush.

Most figures quoted for AI’s slice of web traffic are guesses. Some decks claim AI search already drives 12% to 18% of referrals, but public measurement data points much lower, and that gap matters when you set next quarter’s content budget.

This AI traffic share report pulls together the figures we could verify on the original study pages in September 2026: Statcounter’s monthly referral data, Ahrefs’ 3,000-site AI traffic study, Pew Research’s Google click study, and zero-click research from SparkToro, Semrush and Similarweb. Where we add interpretation, we label it as upGrowth analysis.

What Is the AI Traffic Share in 2026? (Quick Answer)

AI chatbots sent 0.25% of the search, social and AI chatbot referrals Statcounter tracked worldwide in August 2026, up from 0.23% in August 2025. ChatGPT sent 79.4% of those AI referrals. The larger shift sits inside Google, where AI Overviews reduce clicks to websites.

Direct AI referral traffic is still small and hasn’t grown in a straight line, but AI’s influence on whether people click at all is far bigger.

AI traffic statistics: 7 verified findings

1. AI chatbots sent 0.25% of referrals worldwide in August 2026, against 91.66% for search engines and 8.09% for social platforms (Statcounter).
2. The share peaked at 0.43% in April and May 2026 before falling back to 0.25% in August 2026.
3. ChatGPT sent 79.4% of AI chatbot referrals, followed by Google Gemini (10.9%), Perplexity (4.3%), Microsoft Copilot (2.8%) and Claude (2.6%) in August 2026 (Statcounter).
4. Gemini’s share grew from 2.2% to 10.9% between August 2025 and August 2026, while Perplexity’s fell from 8.1% to 4.3%.
5. 63% of websites received at least some AI traffic in Ahrefs’ study of 3,000 sites (February 2025), but AI accounted for only 0.17% of the average site’s visitors.
6. Google users clicked a traditional result in 8% of visits when an AI summary appeared, against 15% without one (Pew Research Center, March 2025 data).
7. AI Overviews correlated with a 58% lower CTR for the top-ranking page, comparing December 2025 with December 2023 (Ahrefs).

AI Traffic Share by Platform: ChatGPT, Gemini, Perplexity and Copilot

ChatGPT dominates AI referral traffic with 79.4% of AI chatbot referrals worldwide in August 2026, according to Statcounter. Gemini is a distant but fast-rising second at 10.9%, and no other platform sends more than 5%.

The table compares each platform’s share of AI chatbot referrals a year apart, plus the August 2026 split in the United States and India. The August 2025 figures come from Statcounter’s downloadable monthly data for the same chart.

ChatGPT: still the biggest AI referrer

ChatGPT’s share barely moved in 12 months, slipping from 80.9% to 79.4%. Its audience is huge: Semrush’s June 2025 AI search study reported ChatGPT weekly active users had passed 800M. But scale doesn’t mean clicks. Similarweb found fewer than 7 in 100 ChatGPT prompts included any source citation as of May 2026.

Google Gemini: the fastest riser

Gemini went from 2.2% of AI chatbot referrals in August 2025 to 10.9% in August 2026, and it’s even stronger in India at 12.7%. In upGrowth’s view, Android and Google account integration explains much of this, so track it for mobile-first audiences.

Perplexity: traffic growth has stalled

If you searched for Perplexity traffic growth, the referral data tells a different story from the hype. Perplexity’s share of AI chatbot referrals was 12.1% in April 2025, fell to 8.1% by August 2025 and reached 4.3% in August 2026. In upGrowth’s experience, it can still send engaged visitors for research-heavy topics.

Microsoft Copilot and Claude

Copilot’s worldwide share dropped from 5.2% to 2.8%, but it holds 8% in the United States, where Microsoft 365 and Edge usage at work is high. Claude rose from 0.9% to 2.6% worldwide and reached 4.4% in the US. For B2B brands, upGrowth treats both as work-research channels rather than consumer ones.

AI Referral Traffic Growth, 2025 to 2026: Is AI Traffic Share Rising?

AI usage is growing fast, but AI referral share isn’t. Statcounter’s worldwide AI chatbot share of referrals moved from 0.23% in August 2025 to a peak of 0.43% in April and May 2026, then settled at 0.25% in August 2026.

AI traffic distribution over time

• August 2025: 0.23%
• November 2025: 0.40%
• January 2026: 0.31%
• April and May 2026: 0.43% (peak)
• August 2026: 0.25%

Regional shares were flatter still. The United States sat at 0.16% in both August 2025 and August 2026, and India at 0.09% in both months. upGrowth note: treat single-month swings with care, since panel-based data can shift as the mix of tracked sites changes.

Usage is growing faster than referrals

Visits to AI tools themselves tell a different story. Similarweb’s August 2026 zero-click analysis, citing its 2026 Generative AI Landscape report, says generative AI platforms pull roughly 9.5 billion visits a month worldwide, up 70% from a year earlier. People are asking AI more questions. They just aren’t clicking out much.

What 3,000 websites saw

Ahrefs’ AI traffic study, published in February 2025, gives a site-level view. Across 3,000 sites, 63% received AI traffic, AI sent 0.17% of the average site’s visitors, and ChatGPT (50%), Perplexity (30.7%) and Gemini (17.6%) together sent 98% of it. The smallest sites, with under 999 monthly visitors, got the highest share at 0.56%.

Where AI search traffic could go next

Forecasts vary widely. Semrush projected in June 2025 that AI search visitors could overtake traditional search visitors by early 2028 for digital marketing and SEO topics. That’s a projection for 1 topic area, not a measured share, and the Statcounter data so far shows a much slower climb.

Google Search Market Share vs AI: How Big Is the Threat From OpenAI and Perplexity?

Google still sent 91.1% of search engine referrals worldwide in August 2026, according to Statcounter. The real threat to website traffic in 2026 isn’t ChatGPT or Perplexity replacing Google. It’s Google answering more queries itself.

Google vs ChatGPT referrals

Statcounter’s Google vs ChatGPT comparison shows that for every 1,000 referrals it counted from either source in August 2026, about 998 came from Google. Google remains the main discovery source for most sites.

How often AI Overviews appear

Semrush’s AI Overviews study, updated in December 2025 with data on 10M+ keywords, found AI Overviews on 6.49% of queries in January 2025, 24.61% in July 2025 and 15.69% in November 2025. Pew Research Center found that 18% of the Google searches in its March 2025 panel produced an AI summary. Learn how to appear in Google AI Overviews if your category is affected.

Zero-Click Searches and AI Overviews: Where the Clicks Go

When an AI summary appears, Google users click far less. Pew Research Center found users clicked a traditional result in 8% of visits with an AI summary, compared with 15% without one, and clicked a link inside the summary in just 1% of visits.

What Pew’s browsing data shows

Pew tracked 900 US adults and 68,879 Google searches in March 2025, and published the results on 22 July 2025. Users also ended their browsing session on 26% of pages with an AI summary, compared with 16% of pages with only traditional results.

AI Overviews and click-through rates

Ahrefs’ February 2026 update compared 300,000 keywords and found AI Overviews now correlate with a 58% lower average CTR for the top-ranking page. Position 1 CTR for AI Overview keywords fell from 0.073 in December 2023 to 0.016 in December 2025. Its April 2025 study had put the drop at 34.5%.

The zero-click baseline

Zero-click search was already the norm before AI Overviews scaled. SparkToro’s 2024 study, using Datos clickstream data, found 58.5% of US Google searches and 59.7% of EU searches ended with zero clicks. Semrush adds nuance: when it compared the same keywords before and after AI Overviews appeared, the zero-click rate actually dipped from 33.75% to 31.53%.

The chatbot zero-click effect

AI chat platforms create the same outcome without a results page. With fewer than 7 in 100 ChatGPT prompts citing a source, most AI answers never produce a visit. upGrowth analysis: that’s why being named and recommended inside the answer matters as much as the click.

Which Industries and Content Types Lose the Most Clicks to AI?

Informational content loses the most, and transactional or local content loses the least. We haven’t found a public dataset that measures AI referral share by industry reliably, so this section is upGrowth analysis based on how AI answers work. For sector detail, see our AI Overviews traffic loss breakdown by vertical.

Industries most exposed

• Health and education: AI answers cover definitions and explainers well, which reduces the need to click.
• Media and publishing: AI summarises the news and reference questions that once drove large traffic volumes.
• Technology and SaaS: high AI adoption among buyers, so well-documented products get cited while thin ones get skipped.
• E-commerce: a split picture, with product research moving into AI while transactional queries still drive organic visits.
• Local services: relatively insulated, because local and real-time queries still need maps, reviews and a visit.

Content types that lose and hold traffic

FAQ and glossary pages, generic listicles, basic how-to guides and news summaries are the easiest for AI to answer in full. Original research, interactive tools, expert analysis and case studies are harder to replace, and AI systems need them as sources.

Winners and losers

In our analysis, the winners are authoritative niche publishers, SaaS companies with strong documentation, data publishers and brands with clear expertise signals. The losers are thin content aggregators, ad-heavy media sites, generic directories and sites that depend on long-tail informational queries AI can answer directly.

AI Traffic Attribution Challenges: Why Analytics Undercount AI

Your analytics show only the AI traffic that arrives with a trackable referrer. Ahrefs notes that some AI traffic shows up as direct, so measured AI traffic is a minimum. Our AI visibility measurement stack guide covers the tooling.

AI Overviews blend into Google organic

Clicks from AI Overviews arrive as ordinary google.com organic traffic. There’s no native split in Google Analytics between a click from an AI Overview and a click from a blue link, so the change shows up as falling CTR rather than a new channel.

Dark AI traffic

Many AI-influenced visits carry no AI referrer at all. A user reads a ChatGPT answer, then types your URL, searches your brand name on Google or clicks a social post. Those visits land in direct, branded organic or social, and last-click attribution gives AI no credit.

Cross-platform journeys and embedded AI

Buyers move between ChatGPT, Gemini, Perplexity and Google in a single journey, and AI features embedded in apps and browsers send visits that look like direct or host-platform referrals.

How to measure AI traffic share in GA4

1. Create a custom channel group for AI referrers such as chatgpt.com, chat.openai.com, perplexity.ai, gemini.google.com, copilot.microsoft.com and claude.ai.
2. Watch Search Console for queries where impressions rise while CTR falls, a common sign that AI answers are absorbing clicks.
3. Track branded search volume and direct traffic to spot an AI halo effect.
4. Ask new leads how they found you, and add AI tools as an answer option.
5. Use UTM parameters on links you control in AI-facing profiles and listings.
6. Build a dashboard that separates AI referrals from traditional organic traffic.

What the AI Traffic Share Data Means for Marketing Leaders

Don’t plan for AI to replace your Google traffic this year, but plan for Google to send fewer clicks per query. The priorities below are upGrowth’s recommendations based on the data in this report.

1. Invest in generative engine optimization now

AI referrals are small, but Semrush found the average AI search visitor was worth 4.4 times the average traditional organic visitor, based on conversion rate. Audit your visibility across ChatGPT, Gemini, Perplexity and AI Overviews, structure pages for citation with clear headings, data and named sources, and build topical authority. Our generative engine optimization services cover this end to end.

2. Diversify traffic sources

Give AI referrals their own targets, and build direct audience relationships through email, communities and first-party data so no single discovery channel controls your pipeline.

3. Rethink content for the AI era

Shift effort from high-volume informational posts to content AI can’t easily reproduce, such as proprietary data, calculators and expert analysis, then update your best pages so AI can quote them accurately.

4. Fix attribution before you scale

Set up the AI channel group, establish a baseline for AI referrals and branded search, and add AI touchpoints to your attribution model. Without a baseline, you can’t tell whether GEO work is paying off.

5. Budget for GEO as a line item

upGrowth recommendation, September 2026: ring-fence 10% to 15% of your SEO and content budget for AI visibility audits, citation-focused content updates, structured data and monitoring. Brands in health, education, finance and technology, where AI answers are most common, should sit at the higher end.

AI Traffic Share: FAQs

Your Next Move: Measure Your Own AI Traffic Share

Industry averages are a starting point, not a target. Your own share depends on your category, your content and how often AI tools cite you.

AI already shapes how often Google sends clicks and which brands get named in answers. Brands that publish data-rich content and measure AI as its own channel are the ones AI keeps citing.

Request an AI visibility audit to see where your brand appears in ChatGPT, Gemini, Perplexity and Google AI Overviews, or contact upGrowth for help with GEO, content optimization and AI attribution setup.

About the Author

Amol Ghemud

Optimizer in Chief

Amol has helped catalyse business growth with his strategic & data-driven methodologies. With a decade of experience in the field of marketing, he has donned multiple hats, from channel optimization, data analytics and creative brand positioning to growth engineering and sales.