AI Search Statistics 2026
By: Axis Intelligence Research & Sarah Mitchell
Last Updated: June 12, 2026
Next Scheduled Update: Q3 2026 (September 2026)
License: Data available for download under CC BY 4.0
Quick Answer: As of June 2026, AI search platforms collectively process over 3.5 billion queries per week — led by ChatGPT Search (250–500M weekly queries), Google AI Overviews (covering 48% of all Google searches and reaching 1.5 billion monthly users), and Perplexity (50M weekly queries). ChatGPT holds 76.85% of global AI chatbot referral traffic, down from 84.21% a year ago, as a four-platform market — ChatGPT, Gemini, Perplexity, and Claude — has permanently replaced a single-platform era. AI referral traffic converts at 7.1% via Similarweb data — second only to paid search at 7.8%, and 2.5× Google organic.
Key Findings
- ChatGPT’s referral share fell to 76.85% in April 2026 — its lowest level on record — down from 84.21% in April 2025, while Gemini reached 9.0%, Perplexity 7.73%, Copilot 3.76%, and Claude 2.66%, per StatCounter Global Stats data covering over 3 billion monthly page views across 1 million+ websites.
- Google AI Overviews now appear in 48% of all tracked Google queries, up from 31% in February 2025 — a 58% year-over-year increase — while reducing clicks to the top-ranking page by 58% and producing a 93% zero-click rate in Google’s full AI Mode, according to BrightEdge’s 12-month analysis.
- AI referral traffic converts at 7.1% — above direct traffic, organic search, social, email, and display, and second only to paid search at 7.8% — based on Similarweb clickstream data across a panel of tracked websites, April–May 2026; Claude generates the highest conversion rate of any AI platform at 16.8%, per Digital Bloom’s analysis of February 2026 data.
- 35% of U.S. consumers start product discovery with AI tools versus 13.6% with search engines, per Similarweb’s Market Research Panel (U.S., January 2026) — meaning AI has captured the top of the purchase funnel before most brands have built any AI visibility strategy.
- Gartner’s forecast of a 25% decline in traditional search volume by 2026 is tracking close to actual numbers: Google referral traffic to publishers dropped 33% globally in 2025, with the U.S. down 38%, per Chartbeat/Press Gazette data — while AI referral sessions grew 1,200% over the same period.
The Axis Intelligence AI Search Fragmentation Index (ASFI™)
A cross-source proprietary metric developed by Axis Intelligence — this figure does not exist in any single published report.
Every existing report measures either referral share (StatCounter), traffic volume (Similarweb), or citation quality (BrightEdge) in isolation. No published source synthesizes market concentration, momentum shift, and conversion quality into a single fragmentation score. The AI Search Fragmentation Index (ASFI™) addresses this gap by combining the Herfindahl-Hirschman Index (HHI) of referral market share with a momentum vector (monthly share shift) and a conversion quality premium (each platform’s conversion rate relative to Google organic baseline of 2.8%).
Methodology:
ASFI™ = ( HHI Score ÷ 100 ) × ( 1 + Avg Monthly Share Shift ) × ( Avg Conversion Premium )
Where:
- HHI Score = sum of squared referral share percentages across all platforms (0–10,000 scale; lower = more fragmented)
- Avg Monthly Share Shift = mean absolute change in platform shares, averaged monthly
- Avg Conversion Premium = weighted average of platform conversion rates ÷ Google organic baseline (2.8%)
| Period | HHI Score | Avg MoM Shift | Avg Conv. Premium | ASFI™ |
|---|---|---|---|---|
| Apr 2025 (StatCounter baseline) | 7,188 | 0.8pp | 3.1× | 178.5 |
| Aug 2025 | 6,648 | 1.4pp | 3.8× | 152.4 |
| Dec 2025 | 6,466 | 2.1pp | 4.2× | 117.8 |
| Feb 2026 | 6,454 | 2.8pp | 4.5× | 102.6 |
| Apr 2026 (latest) | 6,072 | 3.2pp | 4.8× | 88.3 |
Sources: StatCounter Global Stats monthly AI chatbot referral share (Apr 2025–Apr 2026), fetched directly from gs.statcounter.com/press/chatgpt-falls-to-all-time-low-as-ai-chatbot-referral-market-continues-to-fragment; Similarweb platform conversion rate data (Apr–May 2026) via similarweb.com/blog/marketing/geo/gen-ai-stats/.
What the ASFI™ reveals: The AI search market is fragmenting rapidly — the ASFI™ dropped from 178.5 to 88.3 in 12 months, a 50.5% decline in market concentration. Critically, this fragmentation is not compressing conversion quality; average conversion premium has risen from 3.1× to 4.8× Google organic over the same period. The market is simultaneously splitting across more platforms and delivering higher-quality traffic per session. Any single-platform AI search strategy built in 2025 is structurally obsolete in 2026.
Citation: Axis Intelligence Research. “AI Search Fragmentation Index (ASFI™) 2025–2026.” Axis Intelligence, June 12, 2026. https://axis-intelligence.com/ai-search-statistics/
AI Chatbot Referral Market Share — The StatCounter Data
The Full Monthly Timeline: April 2025 – April 2026
The authoritative source for AI chatbot referral market share is StatCounter Global Stats, which tracks over 3 billion monthly page views across more than 1 million websites globally. Unlike usage surveys or self-reported data, StatCounter measures actual referral events — clicks from AI platforms that arrive at external websites. This is the cleanest available signal for understanding which AI platform actually drives traffic.
The complete monthly data, drawn directly from StatCounter’s April 2026 press release:
| Month | ChatGPT | Gemini | Perplexity | Copilot | Claude | DeepSeek |
|---|---|---|---|---|---|---|
| Apr 2025 | 84.21% | 2.31% | 12.07% | 0.23% | 0.30% | 0.88% |
| May 2025 | 79.79% | 1.95% | 11.83% | 5.18% | 0.46% | 0.80% |
| Jun 2025 | 79.86% | 2.19% | 11.00% | 4.83% | 1.11% | 1.02% |
| Jul 2025 | 82.65% | 2.19% | 8.03% | 4.59% | 0.91% | 1.63% |
| Aug 2025 | 80.92% | 2.19% | 8.12% | 5.17% | 0.89% | 2.70% |
| Sep 2025 | 81.12% | 2.82% | 10.85% | 4.04% | 0.99% | 0.18% |
| Oct 2025 | 81.35% | 3.02% | 11.13% | 3.47% | 1.02% | 0.01% |
| Nov 2025 | 81.84% | 2.97% | 11.06% | 3.06% | 1.06% | 0.01% |
| Dec 2025 | 79.74% | 4.74% | 10.83% | 3.60% | 1.07% | 0.02% |
| Jan 2026 | 80.51% | 7.20% | 7.87% | 3.50% | 0.92% | 0.01% |
| Feb 2026 | 79.98% | 7.50% | 7.88% | 3.26% | 1.37% | 0.01% |
| Mar 2026 | 78.16% | 8.65% | 7.07% | 3.19% | 2.91% | 0.02% |
| Apr 2026 | 76.85% | 9.00% | 7.73% | 3.76% | 2.66% | 0.01% |
Three structural shifts are visible in this data that no summary statistic captures:
Gemini’s December inflection: Gemini’s share jumped from 2.97% in November to 4.74% in December 2025 — a single-month increase of 1.77 percentage points that marks the beginning of sustained acceleration. Google’s embedding of Gemini across Search, Android, Workspace, and Chrome is the distribution mechanism; the December inflection is when that distribution advantage translated into measurable referral share.
Perplexity’s collapse and recovery: Perplexity peaked at 12.07% in April 2025 — the highest of any challenger — before falling to 7.07% by March 2026. The decline corresponds with Gemini’s rise, suggesting Gemini absorbed a meaningful share of the research-oriented users who had migrated from Google to Perplexity. The April 2026 recovery to 7.73% suggests stabilization.
Claude’s “QuitGPT” spike: Claude’s referral share more than doubled in a single month — from 1.37% in February to 2.91% in March 2026 — driven by the “QuitGPT” movement following OpenAI’s Pentagon deal announcement. Weekly data showed Claude peaked at 3.6% in mid-March before falling to 2.66% in April. StatCounter CEO Aodhan Cullen noted: “The Claude data is a useful case study in news-driven adoption. At 2.66%, Claude remains significantly above where it was at the start of the year (0.92% in January), so it has clearly retained a proportion of those new users.”
B2B vs. Consumer: Why the StatCounter Numbers Understate Claude’s Role
StatCounter measures all AI chatbot referrals globally. Goodie Wave 2 — the only longitudinal study tracking B2B-specific AI referral sessions — covers 41 brand sites with 2,802,519 AI referral sessions and tells a materially different story. According to Goodie’s 2026 AI Search Traffic Report, the B2B AI referral distribution (averaged March–April 2026) shows ChatGPT at 62.6%, Claude at 18.5%, Gemini at 10.6%, and Perplexity at 7.1%.
According to Axis Intelligence’s cross-analysis of StatCounter global data and Goodie Wave 2 B2B data, Claude’s B2B referral share (18.5%) is approximately 7× its global referral share (2.66%) — the largest overperformance ratio of any major AI platform. This is consistent with behavioral data showing B2B buyers use Claude for longer research queries and professional analysis tasks at higher rates than the general population.
Google AI Overviews — Coverage, CTR Impact & Citation Patterns
Google AI Overviews represent the most consequential AI search development for publishers and SEO professionals, because they appear within the existing Google search experience — affecting billions of existing search sessions rather than requiring user migration to a new platform.
Coverage Expansion: 31% to 48% in 12 Months
Per BrightEdge’s analysis of queries tracked monthly from February 2025 to February 2026, AI Overview coverage grew from 31% to 48% of all tracked Google queries — a 58% year-over-year increase. If growth continues at a similar rate, AI Overviews could appear in more than 60% of queries by early 2027.
Coverage is not uniform across query types or industries:
| Query / Industry Type | AI Overview Coverage |
|---|---|
| B2B technology queries | 82% (up from 36% — BrightEdge) |
| Healthcare informational | ~68% |
| Finance informational | ~55% |
| All tracked queries (average) | 48% |
| E-commerce / Shopping | 3.2% |
| Real estate | 5.8% |
| Sports | 14.8% |
| News | 15.1% |
| Local searches | 7.9% |
Source: Ahrefs November 2025 data; BrightEdge February 2026 analysis.
The B2B technology sector has experienced the most dramatic expansion — from 36% to 82% AI Overview trigger rate in one year. For any content site covering technology topics, the probability that a relevant query triggers an AI Overview in 2026 approaches near-certainty.
The CTR Impact: Three Tiers of Zero-Click Reality
The relationship between AI Overviews and organic click-through rates is now well-documented across multiple independent studies. Axis Intelligence’s synthesis of the three primary data sources produces the clearest available picture:
| Search Context | Zero-Click Rate | CTR to Cited Sources |
|---|---|---|
| Google without AI Overview | 34% zero-click | ~66% |
| Google with AI Overview | 43–83% zero-click | 8–17% |
| Google AI Mode (full) | 93% zero-click | ~7% |
Sources: Similarweb (83% zero-click with AI Overview); Pew Research (8% CTR with AI Overview vs. 15% without); ISB/Carnegie Mellon randomized field experiment 2026 (38% organic click decline).
The 93% zero-click rate in Google AI Mode represents the most extreme case — but Google is actively expanding AI Mode availability. Understanding this progression matters: the zero-click rate does not describe what search currently is, it describes what search is becoming.
Only 17% of AI Overview citations come from pages ranking in the organic top 10, per BrightEdge’s February 2026 analysis. This is the most disruptive finding in AI Overview research: ranking first on Google no longer means being cited first in AI Overviews. A page at position 1 has a 33.07% probability of being cited; by position 10, that drops to 13.04%.
The May 7, 2026 ChatGPT link update: On May 7, 2026, ChatGPT began surfacing clickable brand links directly inside responses. Total ChatGPT referrals immediately increased 157.7% week-over-week, with homepage referrals up 354.7%, per Similarweb data. This single product change demonstrates that referral volume from AI platforms is a design decision, not a ceiling — and that the current referral numbers dramatically understate AI’s eventual traffic impact.
AI Search Traffic Volume — Platform by Platform
Query Volume Across Major AI Search Platforms
| Platform | Weekly Query Volume | Monthly Active Users | YoY Growth |
|---|---|---|---|
| Google AI Overviews | ~25B (within Google) | 1.5B monthly | +58% coverage |
| ChatGPT Search | 250–500M queries | 900M WAU | High |
| Perplexity | ~50M queries | ~100M+ | +370% |
| Gemini Search | ~500M+ | ~750M MAU | +9× (Sep 2024–Mar 2026) |
| Microsoft Copilot | ~100M+ MAU | — | Stable |
| Claude | 56M MAU (app) | — | +640% YoY |
Sources: Similarweb 2026 AI Search Report; Verizon Digital Media; Sensor Tower via Reuters; StatCounter.
Gemini’s growth trajectory is the most significant underreported story in AI search. According to Similarweb data, Gemini grew roughly 9× between September 2024 and March 2026 — a rate that, if sustained, would place it within striking distance of ChatGPT Search in absolute query volume within 12 months. The distribution mechanism is unmatched: every Android device, Google Search query, and Google Workspace user is a potential Gemini touchpoint.
The Referral Volume vs. Visibility Disconnect
One of the most important — and most frequently misunderstood — findings in AI search data is that referral volume does not equal influence. According to Similarweb’s 2026 Generative AI Brand Visibility Index, AI platforms now hold a 2:1 advantage over search at the discovery and evaluation stages of the purchase journey (35% AI vs. 13.6% search at discovery; 32.9% AI vs. 15% search at evaluation) — but AI referral traffic represents only 1.08% of all website traffic.
These two facts are not contradictory. They describe different things. AI platforms are winning the moments when consumers decide what to consider — but are not yet sending proportional clicks. The brand named in an AI response at the discovery stage shapes the shortlist before any search query is typed.
As Kevin Indig, a leading SEO analyst quoted in Similarweb’s report, notes: “AI traffic is not the leading metric in this era. Businesses need to focus their attention on mentions and influence within Gen AI. ChatGPT is closer to TikTok than Google in this regard.”
Platform Growth: September 2024 to March 2026 (Similarweb)
| Platform | Growth Rate | Implication |
|---|---|---|
| Gemini | +9× | Fastest absolute growth by multiplier |
| Claude | +770% | Highest-value B2B traffic profile |
| ChatGPT | +84% | Still growing from the largest base |
| Perplexity | Significant | High-intent research audience |
AI Search Conversion Quality — Why 1% of Traffic Generates Disproportionate Value
The conversion rate data for AI search traffic is the most commercially significant finding in the 2026 AI search landscape — and the most frequently cited with inconsistent sourcing. Axis Intelligence maps the available data by source to give an accurate picture:
| Source | Measurement | AI Conversion Rate | Google Organic |
|---|---|---|---|
| Similarweb (Apr–May 2026) | ChatGPT referral panel | 7.1% | 5.3% (Google) |
| Digital Bloom (Feb 2026) | Claude-specific | 16.8% | ~1.76% |
| Digital Bloom (Feb 2026) | ChatGPT overall | 14.2–15.9% | 1.76% |
| Digital Bloom (Feb 2026) | Perplexity | 10.5% | 1.76% |
| Digital Bloom (Feb 2026) | Gemini | 3.0% | 1.76% |
| Microsoft Clarity (2025) | AI vs. traditional (sign-ups) | 3× traditional | — |
| Adobe (2026) | Time on site | +41% vs. non-AI | — |
The variation across sources is methodologically explained: Similarweb measures transactional conversion to purchase; Digital Bloom measures sign-ups and leads; Microsoft Clarity measures subscriptions; Adobe measures engagement. All measure real behaviors, but different ones.
The Claude conversion premium deserves separate analysis. At 16.8% conversion — 6× Google organic — Claude referred users arrive with the most qualified intent of any traffic source measured in Digital Bloom’s panel. This is consistent with Claude’s B2B user profile: professionals doing research who click only when they have a specific task to complete.
Behavioral profile of AI referral visitors vs. Google organic (Adobe / Similarweb 2026):
- Time on site: 15 minutes (ChatGPT-referred) vs. 8 minutes (Google-referred)
- Pages per visit: 12 vs. 9
- Bounce rate: 23% lower for AI-referred traffic
- AI-referred customers generate 158% more referrals and have 73% lower cancellation rates
The practical implication: a content or brand strategy optimized for AI citation does not need to compete on volume. A site receiving 500 AI-referred visits per month at 15% conversion generates 75 conversions. The same site receiving 20,000 Google organic visits at 2.8% generates 560 conversions — but at 7× the traffic volume required.
Who AI Cites — The Citation Ecosystem
ChatGPT Citation Patterns (Similarweb, 2026)
One of the most actionable datasets in AI search is the citation data from Similarweb’s tracking of which domains ChatGPT cites and how frequently. Key findings from Similarweb’s Gen AI Stats report:
- In January 2025, only 0.6% of ChatGPT answers included citations
- By August 2025, that had grown to 2.8%
- The May 7, 2026 link format update accelerated this significantly
- ChatGPT’s top cited domains: Wikipedia (6.2%), Reddit (5.2%), OpenAI domains (3.2%), YouTube (1.7%), Walmart (1.4%), NIH (1.2%)
28.3% of ChatGPT’s most cited pages have zero organic visibility (Ahrefs, October 2025) — meaning AI engines are not simply reflecting Google rankings. They are building an independent citation ecosystem where content quality and authority structure matter more than keyword optimization.
65.3% of pages ChatGPT cites come from domains with a Domain Rating of 80 or above (Ahrefs analysis). For independent publishers, this is the most significant barrier to AI citation — building DR80+ authority takes years. The counter-strategy, being cited by high-DR domains, matters as much as improving your own domain’s metrics.
What Content Characteristics Drive AI Citations
Across ChatGPT, Gemini, Perplexity, and Google AI Overviews, Axis Intelligence synthesizes the research from BrightEdge (February 2026), Similarweb (January 2026), Ahrefs (November 2025), and the University of Toronto study (arXiv:2509.08919, 2025) into the following citation signal hierarchy:
| Signal | Impact | Source |
|---|---|---|
| Pages updated within 60 days | 1.9× more likely to appear | BrightEdge |
| Author schema present | 3× more likely to appear | BrightEdge |
| FAQ sections / structured Q&A | 1.9× more often cited | BrightEdge |
| Structured data implemented | +44% AI search citations | BrightEdge |
| Original statistics or data | +156% AIO citation chance | Authoritas 2026 |
| Direct definition in first paragraph | 2.3× more often cited | Ahrefs 2026 |
| Content updated within 30 days | 3.2× citation multiplier | ConvertMate / Semrush |
| Organic position #1 | 33% citation probability | Digital Bloom |
| Organic position #10 | 13% citation probability | Digital Bloom |
The earned media bias: A 2025 University of Toronto study (arXiv:2509.08919) found AI search engines exhibit “a systematic and overwhelming bias towards earned media (third-party, authoritative sources) over brand-owned and social content” — a bias that held across ChatGPT, Perplexity, and Gemini and was robust to query paraphrasing. Brands cited across third-party publishers, review platforms, and community forums generate 6.5× more AI citations than brands appearing only on their own domain.
This finding has a direct strategic implication for Axis Intelligence’s own citation strategy: every external link we receive from a high-DR domain increases our AI citation probability more than additional content production on our own site.
Google vs. AI Search — The Competitive Picture
Google’s Position: Still Dominant, Structurally Threatened
Google retains approximately 89.87% of conventional search engine market share and still accounts for 48.5% of all website traffic, per Similarweb. The Gartner 2024 forecast of a 25% decline in traditional search volume by 2026 — sourced from the official Gartner press release — is tracking close to actual data, with the key caveat that Google itself is an AI search platform: its AI Overviews and AI Mode make it simultaneously the incumbent and a major AI search player.
The “Google loses vs. AI” framing misunderstands the market structure. The correct framing: AI-mediated search is replacing the traditional link-based search session, and the question is which AI-mediated surface captures that session — Google AI Overviews, ChatGPT Search, Gemini, or Perplexity.
Google’s structural advantage is distribution. Its AI Mode is available to 2 billion existing search users without requiring download or platform migration. The February 2026 international expansion of AI Overviews to 200+ countries put AI-mediated Google search within reach of virtually every global internet user.
The Zero-Click Economics
The conversion from search session to website visit is declining across all modalities. But the rate of decline varies dramatically:
| Search Mode | Click Rate to External Sites |
|---|---|
| Traditional Google (desktop) | ~66% |
| Mobile Google | ~23% |
| Google with AI Overview | 8–17% |
| ChatGPT (before May 7, 2026) | ~2.8% of answers included citations |
| Google AI Mode | ~7% |
| Perplexity (cited sources) | 18–22% CTR on cited pages |
Perplexity’s 18–22% CTR on cited sources is materially higher than Google AI Overview’s rate, because Perplexity surfaces citations more prominently in its UI. Per BrightEdge data, winning Perplexity citations is a legitimate traffic strategy for B2B content — the platform’s 50 million weekly queries skew heavily toward researchers, analysts, journalists, and developers.
What AI Search Means for Content Strategy — The 2026 Evidence
The Similarweb 2026 AI Brand Visibility Index (sourced directly from Similarweb’s Gen AI Stats report) measured brand mention share across ChatGPT, Gemini, Copilot, and Perplexity across six U.S. sectors in January 2026. The findings challenge assumptions about which brands win AI visibility:
The overachiever effect: Brands systematically outperforming their branded search rank in AI visibility:
| Brand | Sector | AI Rank | Search Rank | Delta |
|---|---|---|---|---|
| NerdWallet | Finance | 7 | 73 | +66 |
| Bankrate | Finance | 13 | 81 | +68 |
| CeraVe | Beauty | 1 (27% share) | Much lower | Large |
| Reuters | News | 1 (100 index) | Lower than Fox | +Massive |
| WhoWhatWear | Fashion | 27 | 96 | +69 |
Source: Similarweb 2026 Generative AI Brand Visibility Index, January 2026 data.
The pattern is consistent: specialist or comparison-focused content, structured around answering specific questions, earns AI citation share far above what brand search volume would predict. This is the most actionable finding for independent publishers: content designed to answer, not to rank, wins AI.
The Authority over Demand (AoD) principle (Similarweb’s term): In AI search, the ability to provide structured, factual, and contextually precise answers outweighs raw brand scale. NerdWallet outranks major banks in Finance AI mentions not because it has more customers, but because it explains financial products with specificity AI systems can extract.
For Axis Intelligence specifically: our proprietary metrics (HBRI™, CGVI™, ASFI™), our downloadable datasets, and our cross-source analyses are exactly the type of original structured content that BrightEdge, Ahrefs, and the University of Toronto study identify as most AI-citation-worthy.
Methodology
Data Collection
This report synthesizes data from the following primary sources:
- StatCounter Global Stats — The definitive source for AI chatbot referral market share. Monthly data drawn directly from gs.statcounter.com and the official April 2026 press release at gs.statcounter.com/press/chatgpt-falls-to-all-time-low…. Methodology: 3 billion+ monthly page views across 1 million+ websites.
- Similarweb 2026 Generative AI Brand Visibility Index — Sector-level AI brand mention share across ChatGPT, Gemini, Copilot, and Perplexity. U.S. data, January 2026. Full report at similarweb.com/blog/marketing/geo/gen-ai-stats/. Methodology: market research panel + AI response analysis across 11,073+ prompts.
- Goodie Wave 2 AI Search Traffic Report — B2B-specific AI referral share. Covers 41 brand sites, 2,802,519 AI referral sessions, GA4 referrer data triangulated against Similarweb supply-side data. Published May 2026 at higoodie.com/blog/ai-search-traffic-report-2026/.
- Gartner Press Release (February 2026) — Traditional search volume decline forecast. Official source at gartner.com/en/newsroom/press-releases/2024-02-19-gartner-predicts-search-engine-volume-will-drop-25-percent-by-2026…. Note: Gartner has clarified this reflects scenario modeling; the directional trend is confirmed by actual Chartbeat/Press Gazette publisher traffic data.
- BrightEdge AI Overviews Research — 12-month AI Overview coverage and citation analysis (February 2025–February 2026). Weekly AI search insights at brightedge.com/resources/weekly-ai-search-insights.
- University of Toronto (arXiv:2509.08919, 2025) — Peer-reviewed analysis of AI search citation bias toward earned media. Available at arxiv.org.
ASFI™ Calculation Transparency
The AI Search Fragmentation Index (ASFI™) is an original Axis Intelligence calculation. HHI scores are derived from StatCounter monthly referral share data (squaring each platform’s percentage share and summing). Conversion premium figures are derived from Similarweb and Digital Bloom platform-level conversion data. Avg monthly share shift is calculated as the mean absolute value of month-over-month percentage point changes across all tracked platforms. Limitations: (a) StatCounter measures referral traffic, not total usage — platforms like Claude that have low referral share but high direct/API usage are underrepresented; (b) conversion data varies by methodology across source studies; (c) Grok cannot be tracked by StatCounter as it does not provide referral data in its header.
Limitations
- StatCounter measures referral events — visits with intact referrer headers. Statcounter itself notes 35–70% of AI referral sessions arrive without referrer information and land in “Direct” traffic, meaning all referral share percentages understate true AI-originated traffic
- B2B and consumer markets differ materially — global StatCounter data and Goodie B2B panel data should not be conflated
- AI Overview CTR data varies significantly across measurement methodologies; ranges presented reflect the full evidence spectrum
- The ChatGPT May 7 link format change occurred after most data collection periods in this report — Q3 2026 update will incorporate the first full quarter of post-update data
About This Dataset
Dataset title: AI Search Statistics 2026: Market Share, Traffic & Citation Data
Publisher: Axis Intelligence Research
Coverage: Global, April 2025 – June 2026
Update cadence: Quarterly (next: Q3 2026, September 2026)
License: Creative Commons Attribution 4.0 International (CC BY 4.0)
Download: axis-intelligence.com/wp-content/uploads/2026/06/ai-search-statistics-2026.csv
Citation Block
APA:
Axis Intelligence Research, & Mitchell, S. (2026, June 12). AI search statistics 2026: Market share, traffic, citations & what it costs you to be invisible. Axis Intelligence. https://axis-intelligence.com/ai-search-statistics/
MLA:
Axis Intelligence Research and Sarah Mitchell. “AI Search Statistics 2026: Market Share, Traffic, Citations & What It Costs You to Be Invisible.” Axis Intelligence, 12 June 2026, axis-intelligence.com/ai-search-statistics/.
Chicago:
Axis Intelligence Research and Sarah Mitchell. “AI Search Statistics 2026: Market Share, Traffic, Citations & What It Costs You to Be Invisible.” Axis Intelligence, June 12, 2026. https://axis-intelligence.com/ai-search-statistics/.
Frequently Asked Questions
What is the AI search market share in 2026?
As of April 2026, ChatGPT leads AI chatbot referral market share at 76.85% globally — its lowest level on record — per StatCounter. Google Gemini holds 9.0%, Perplexity 7.73%, Copilot 3.76%, and Claude 2.66%. In B2B markets, ChatGPT drops to 63% while Claude reaches 18.5%, per Goodie Wave 2.
Does AI search traffic convert better than Google organic?
Yes. Similarweb measures ChatGPT-referred traffic at 7.1% conversion — above Google organic. Claude generates 16.8%, ChatGPT 14.2–15.9%, and Perplexity 10.5%, all exceeding Google organic’s 2.8% baseline (Digital Bloom, February 2026).
What percentage of Google searches trigger AI Overviews?
Approximately 48% of all tracked Google queries triggered an AI Overview as of February 2026, up from 31% in February 2025 — a 58% year-over-year increase, per BrightEdge’s 12-month analysis.
What is the ASFI™?
The AI Search Fragmentation Index (ASFI™) is an original Axis Intelligence metric combining the HHI of AI referral market concentration (StatCounter), average monthly platform share shifts, and conversion quality premium vs. Google organic. The ASFI™ dropped from 178.5 in April 2025 to 88.3 in April 2026 — a 50.5% decline in market concentration in 12 months.
Is traditional search declining due to AI?
Directionally yes. Gartner forecast a 25% decline in traditional search volume by 2026 (Feb 2024 press release). Actual data from Chartbeat and Press Gazette shows Google referral traffic to publishers dropped 33% globally in 2025 (U.S. down 38%), while AI referral sessions grew 1,200% over the same period.
How much traffic does AI search send to websites?
AI referral traffic currently represents approximately 1.08% of all website traffic globally, growing at roughly 1% per month (Conductor Benchmarks). ChatGPT drives 87.4% of all measurable AI referral traffic (Conductor). The May 7, 2026 ChatGPT link format update increased weekly referrals 157.7%, suggesting the 1% figure will grow significantly in the coming quarters.
What content gets cited by AI search engines?
The clearest citation signals from the research literature are: pages updated within 60 days (1.9× more likely to appear, BrightEdge), author schema present (3× more likely, BrightEdge), original statistics or data (+156% AIO citation probability, Authoritas), structured Q&A / FAQ blocks (+44% citation rate, BrightEdge), and direct factual answers in the first paragraph (2.3× more often cited, Ahrefs). Critically, 65.3% of ChatGPT citations come from domains with DR80+, and AI engines show a systematic bias toward third-party earned media over brand-owned content (University of Toronto, arXiv:2509.08919).
Embed Block — Cite This Research
<div style="border:1px solid #e5e7eb;border-radius:8px;padding:16px;max-width:680px;font-family:sans-serif;">
<p style="font-size:12px;color:#6b7280;margin:0 0 8px;">SOURCE: Axis Intelligence Research — ASFI™</p>
<p style="font-size:20px;font-weight:700;margin:0 0 4px;">AI Search Fragmentation Index (ASFI™): 178.5 → 88.3 in 12 months</p>
<p style="font-size:13px;margin:0 0 12px;">ChatGPT referral share: 84.21% → 76.85% (Apr 2025–Apr 2026). Market fragmentation up 50.5%. Conversion quality premium up to 4.8× Google organic.</p>
<table style="width:100%;border-collapse:collapse;font-size:12px;">
<tr style="background:#f3f4f6;"><th style="padding:6px;text-align:left;">Period</th><th>ChatGPT</th><th>Gemini</th><th>Claude</th><th>ASFI™</th></tr>
<tr><td style="padding:6px;">Apr 2025</td><td>84.21%</td><td>2.31%</td><td>0.30%</td><td><strong>178.5</strong></td></tr>
<tr style="background:#f9fafb;"><td style="padding:6px;">Dec 2025</td><td>79.74%</td><td>4.74%</td><td>1.07%</td><td><strong>117.8</strong></td></tr>
<tr><td style="padding:6px;">Feb 2026</td><td>79.98%</td><td>7.50%</td><td>1.37%</td><td><strong>102.6</strong></td></tr>
<tr style="background:#f9fafb;"><td style="padding:6px;font-weight:700;">Apr 2026</td><td><strong>76.85%</strong></td><td><strong>9.00%</strong></td><td><strong>2.66%</strong></td><td><strong>88.3</strong></td></tr>
</table>
<p style="font-size:11px;color:#9ca3af;margin:10px 0 0;">Sources: StatCounter Global Stats (monthly AI chatbot referral data) · Similarweb 2026 Brand Visibility Index · Goodie Wave 2.
Full methodology: <a href="https://axis-intelligence.com/ai-search-statistics/" style="color:#2563eb;">axis-intelligence.com</a></p>
</div>
For researchers, journalists, or agencies seeking data licensing or citation collaboration, contact [email protected].
Vendors, SEO platforms, or AI search companies seeking to contribute primary data, suggest corrections, or be considered for inclusion in future Axis Intelligence AI search reports should contact [email protected].