AI SEO Statistics 2026: Adoption, AI Overviews & LLM Citation Data

An estimated 86% of SEO professionals have integrated AI into their workflow, according to Aira’s 2025 State of SEO survey of 2,500 practitioners. At the same time, Google’s AI Overviews now appear in 13.14% of all searches as of March 2025, more than double the 6.49% recorded in January 2025, per a Semrush analysis of more than 10 million keywords. The two numbers describe the same shift from opposite sides of the screen: SEOs are racing to use AI while AI quietly rewrites the search experience their content depends on.

The pressure is measurable. Ahrefs found that position 1 organic click-through rate fell from 1.41% to 0.64% on pages where an AI Overview appears, based on a 300,000-keyword study published in 2025. Pew Research Center monitored the browsing of 900 US adults in March 2025 and found click rates on traditional results dropped to 8% on AI Overview pages, compared with 15% on pages without an AI summary. The data below covers AI tool adoption, AI Overview impact, LLM citation behavior, AI-generated content prevalence, and publisher traffic shifts through Q1 2026.

Key Takeaways

• 86% of SEO professionals have integrated AI into their SEO strategy, with the average enterprise team using 4.2 different AI tools.
• AI Overviews appear in 13.14% of Google searches as of March 2025, more than doubling from 6.49% in January 2025.
• Position 1 organic CTR dropped from 1.41% to 0.64% when an AI Overview is present, an effective 54% click loss.
• An estimated 74% of newly created web pages in April 2025 contained AI-generated content, per an Originality.ai analysis of 65,000 URLs.
• Only approximately 20% of URLs cited by ChatGPT and Perplexity also rank in Google’s top 10 for the same query, according to BrightEdge’s Generative Parser dataset.
• 58.5% of US Google searches resulted in zero clicks in 2024, with the European Union rate approximately 59.7%, per SparkToro’s annual study by Rand Fishkin.
• 78% of enterprise SEO teams now use AI for keyword research, the most-adopted AI use case across the SEO workflow.

Editor’s Choice

• AI Overviews trigger on 88% of health-related queries and 67% of “people and society” queries, coverage that more than doubled between January and March 2025.
• Pew Research found users were nearly twice as likely to end their browsing session entirely after a search page with an AI Overview.
• Google AI Overviews reached over 100 countries and 30 languages by October 2024.
• An estimated 42% of SEOs say AI tools have substantially or entirely replaced traditional keyword research tools, with the rate climbing to 58% at organizations with over 500 employees.
• Reddit, Wikipedia, and YouTube combined account for approximately 24% of all sources cited by ChatGPT, with listicle and how-to formats representing over 40% of LLM citations overall.
• HuffPost’s organic search traffic fell approximately 65% from its 2022 peak by mid-2024.
• 67% of small businesses use AI for content marketing and SEO, per HubSpot’s 2025 State of Marketing report.

Recent Developments

• April 2026, AI Mode anniversary tracker: Google launched AI Mode on March 5, 2025, as an experimental Search experience powered by a custom Gemini 2.0 model.
• March 2026, AI Mode general availability: Google’s AI Mode rolled out to all US users in May 2025.
• February 2026, ChatGPT Search Free-tier milestone: In February 2025, OpenAI extended ChatGPT Search to all users, including the Free tier, following the October 31, 2024, initial launch.
• March 2026, NYT copyright case status: The New York Times v. Microsoft and OpenAI copyright case (SDNY 1:23-cv-11195) remains active in discovery after the April 2025 partial denial of OpenAI’s motion to dismiss.
• February 2026, Pew Research methodology cited downstream: Pew Research’s March 2025 finding that click-through dropped from 15% to 8% on AI Overview pages continues to anchor independent measurement of AI Overview click impact.
• April 2026, Semrush AI Overview tracking baseline: Semrush’s March 2025 baseline shows AI Overviews appearing in 13.14% of queries, more than double the 6.49% January 2025 figure.

AI Adoption Among SEO Professionals

• An estimated 86% of SEO professionals have integrated AI into their SEO strategy, per Aira’s 2025 State of SEO Report covering 2,500 respondents.
• 75% of marketers using AI report that it helps reduce time spent on manual or repetitive tasks, according to HubSpot’s 2025 State of Marketing Report.
• 67% of small businesses use AI for content marketing and SEO, per HubSpot.
• An estimated 65% of businesses report better SEO results after integrating AI tools, per Search Engine Journal’s 2025 productivity survey.
• AI-assisted content production cuts average article production time by 40 to 50%, per Search Engine Journal.
• The average enterprise SEO team uses 4.2 different AI tools in its workflow, per Aira.
• AI strategy adoption among SEOs rose from 65% in 2024 to 86% in 2025, a 21-point year-over-year jump, per Aira.

Source: Aira, HubSpot, Search Engine Journal

By the numbers: According to Aira’s 2025 State of SEO Report, 86% of SEO professionals now integrate AI into their workflow, up from 65% in 2024. The 21-point jump in twelve months is the steepest single-year tooling shift documented in the report’s six-year history, suggesting AI moved from experiment to default workflow component.

For context on the overall search behavior shifts that AI tools are responding to, see our Google search behavior statistics.

Subscribe To Our Newsletter!

Be the first to get exclusive offers and the latest news.

Email address

Subscribe Now

AI Overview Search Coverage by Query Type

• AI Overviews appear in 13.14% of all Google searches as of March 2025, per Semrush’s analysis of more than 10 million keywords.
• AI Overview coverage more than doubled from 6.49% in January 2025 to 13.14% in March 2025.
• Health-related queries trigger AI Overviews 88% of the time, more than double the all-query average, per Semrush.
• People and society queries trigger AI Overviews 67% of the time, in a Semrush sample of more than 10 million keywords.
• By October 2024, AI Overviews had expanded to over 100 countries and 30 languages, per Google’s official announcement.
• AI Overviews launched in the United States in May 2024 and were extended to the United Kingdom, India, Japan, Brazil, and Indonesia within the first six months.

Source: Semrush AI Overviews Study, Google Blog

Click-Through Rate Impact of AI Overviews

• Position 1 organic CTR fell from 1.41% to 0.64% when an AI Overview appears above results, per Ahrefs’ 300,000-keyword study.
• The average informational query CTR dropped 34.5% on pages where AI Overviews appear, per Ahrefs.
• Transactional queries showed less than 10% CTR loss when AI Overviews appeared, indicating asymmetric impact across query types.
• Pew Research found Google users clicked a traditional search result link in only 8% of visits to results pages with AI summaries, compared with 15% of visits to pages without AI summaries.
• Users were nearly twice as likely to end their browsing session entirely after a search page with an AI Overview, per Pew Research’s monitoring of 900 US adults in March 2025.

Key finding: Ahrefs analyzed 300,000 keywords and found position 1 organic CTR fell from 1.41% to 0.64% when an AI Overview appears, an effective 54% click loss for the top result. The damage is concentrated on informational queries (down 34.5%), while transactional queries showed less than 10% CTR loss.

Zero-Click Search Trend

• An estimated 58.5% of US Google searches resulted in zero clicks in 2024, with the EU rate approximately 59.7%, per SparkToro’s study by Rand Fishkin.
• In the European Union, the zero-click rate was approximately 59.7%, slightly higher than the US figure.
• For every 1,000 US Google searches, only approximately 360 clicks went to non-Google properties; 304 clicks stayed on Google-owned properties.
• The zero-click rate has trended upward in correlation with the AI Overview rollout since May 2024, reaching approximately 58.5% of US searches.
• Pew Research’s independent measurement found that users ended their session after viewing a search page with an AI summary at nearly double the rate of pages without one.

Source: SparkToro 2024 Zero-Click Search Study, Pew Research Center

AI-Generated Content on the Web

• An estimated 74% of newly created web pages in April 2025 contained AI-generated content, per Originality.ai’s analysis of 65,000 newly indexed URLs.
• The share of AI-generated articles on the web roughly doubled between 2023 and 2024, rising from approximately 5% to over 10% by some Graphite measures.
• By some Graphite measures, AI-generated content rose roughly two-fold, from approximately 5% of new content in early 2023 to over 10% by late 2024.
• AI-generated articles cluster on lower-domain-authority sites and on topics with high commercial intent, per Originality.ai.
• Growth in AI content production slowed in 2025 as Google’s HCU and core spam updates reduced the SEO ROI of mass AI publishing, after the share roughly doubled (from approximately 5% to over 10%) the prior year, per Graphite.

Source: Originality.ai, Graphite

For a broader context on generative AI adoption beyond search, see our generative AI adoption data.

Google’s Response to AI Content

• Google’s March 2024 core update introduced refined spam policies targeting scaled content abuse, expired domain abuse, and site reputation abuse.
• The March 2024 update reduced low-quality, unoriginal content in Google search results by an estimated 45%, exceeding the initial 40% reduction target.
• Sites producing scaled AI content without original value were disproportionately affected by the March 2024 update, per Google.
• Sites affected by the September 2023 Helpful Content Update lost an average of 70% of organic search visibility, with under 15% recovering by mid-2024, per Sistrix analysis.
• Under 15% of HCU-affected sites had recovered prior visibility by mid-2024, per Sistrix.
• The HCU-affected cohort included a disproportionate share of sites identified as producing mass AI content with low E-E-A-T signals, with under 15% recovering visibility, per Sistrix.

Across our 80+ cybersecurity statistics pages and adjacent AI coverage, the same pattern keeps surfacing in search data: enforcement updates produce sharp visibility cliffs that mass-produced content rarely recovers from, regardless of whether the underlying production method is AI or human.

LLM Citation Patterns vs Google Rankings

• Approximately 20% of URLs cited by ChatGPT and Perplexity also appear in Google’s top 10 results for the same query, per BrightEdge’s Generative Parser dataset.
• The remaining approximately 80% of LLM citations come from URLs that rank below position 10 in Google or do not rank at all, per BrightEdge.
• BrightEdge concludes that LLM optimization requires distinct strategies from traditional Google ranking optimization, with only approximately 20% URL overlap with Google’s top 10.
• Pages with structured data, clear headings, and explicit Q&A sections are over-represented in LLM citations relative to their share of organic search traffic, per Profound.
• Perplexity cites the same source twice as often as ChatGPT does within multi-result answers, per Profound.

Source: BrightEdge, Profound

Key finding: BrightEdge’s Generative Parser dataset shows that approximately 20% of URLs cited by ChatGPT and Perplexity also appear in Google’s top 10 results for the same query. The implication is structural: ranking on Google and earning AI citations are increasingly two separate disciplines, and a content team that wins one is not guaranteed to win the other.

The next section names the sources that earn the most AI citations.

Top Sources Cited by AI Search Engines

• Reddit, Wikipedia, and YouTube combined account for approximately 24% of all sources cited by ChatGPT, with listicle and how-to formats accounting for over 40% of LLM-cited content overall, per Semrush’s analysis of 80 million AI search queries.
• Perplexity skews more heavily toward news publications and academic sources in its citations, while ChatGPT pulls approximately 24% of sources from Reddit, Wikipedia, and YouTube combined, and listicle formats account for over 40% of LLM citations, per Semrush.
• Across 80 million AI search queries spanning ChatGPT, Perplexity, and Google AI Mode, Semrush found user-generated content platforms over-represented relative to traditional Google rankings, contributing approximately 24% of ChatGPT’s citation mix.
• Pages with structured data and explicit Q&A sections show stronger LLM citation rates relative to their organic search share, per Profound.

Source: Semrush LLM Citation Analysis (80M queries)

For deeper coverage of one major SEO research platform, see our Ahrefs platform statistics.

Content Format Preferences in LLM Citations

• Listicle and how-to format pages account for over 40% of cited content across LLM search engines, with Reddit, Wikipedia, and YouTube combined representing approximately 24% of ChatGPT’s source mix, per Semrush.
• The average length of LLM-cited pages ranges from 1,400 to 2,400 words, depending on query type, per Profound.
• LLM-cited pages typically include clear headings, bullet lists, and structured Q&A sections, per Profound’s analysis.
• Across 80 million queries, content with explicit numbered or bulleted answers showed citation rates significantly above the baseline, with listicle and how-to formats capturing over 40% of LLM citations and Reddit, Wikipedia, and YouTube approximately 24% of ChatGPT’s mix.

Source: Semrush, Profound

AI Tool Adoption by SEO Workflow Task

• An estimated 78% of enterprise SEO teams use AI for keyword research, the most-adopted use case, per Conductor’s 2025 State of Organic Marketing survey.
• Content brief generation is the second-most-adopted AI task at 71%, per Conductor.
• Meta description generation, content outlines, and competitor analysis round out the top five AI use cases in enterprise SEO, per Conductor.
• An estimated 42% of SEOs say AI tools have substantially or entirely replaced traditional keyword research tools in their workflow, with the rate climbing to 58% at organizations with over 500 employees, per BrightEdge.
• At organizations with over 500 employees, 58% report substantial AI tool replacement of legacy SEO software, per BrightEdge.

AI in Enterprise SEO Spend

• The average enterprise SEO team uses 4.2 different AI tools in its workflow, per Aira’s 2025 State of SEO Report.
• Content optimization, keyword research, and content brief generation are the most common AI use cases at the enterprise level, per Aira.
• Larger organizations (over 500 employees) show 58% substantial AI replacement of legacy SEO tooling, compared with 42% across all sizes, per BrightEdge.
• 67% of small businesses use AI for content marketing and SEO, per HubSpot.
• Article production time falls by 40 to 50% when AI is integrated into the content workflow, per Search Engine Journal.

Publisher Traffic Impact from AI Search

• News publishers saw an estimated 30 to 40% drop in Google referral traffic between 2023 and 2025, per Press Gazette’s analysis of Similarweb data.
• HuffPost organic search traffic fell approximately 65% from its 2022 peak by mid-2024, per Originality.ai’s publisher impact analysis.
• The publisher traffic decline accelerated after the May 2024 AI Overviews rollout, per Press Gazette.
• Business Insider and several other large digital-first publishers reported organic search traffic declines in excess of 50% from 2022 peaks, per Press Gazette.
• Originality.ai’s analysis correlates the approximately 65% HuffPost decline with both Google’s HCU rollout and the AI Overviews introduction on news-adjacent queries.

Source: Press Gazette, Originality.ai

Publisher losses align with the rapid AI search platform growth covered next.

AI Search Engine User Growth and Adoption

• Google launched AI Mode on March 5, 2025, powered by a custom version of Gemini 2.0, initially for Google One AI Premium subscribers via Search Labs.
• In May 2025, Google expanded AI Mode to all users in the United States, removing the Labs sign-up requirement.
• OpenAI launched ChatGPT Search publicly on October 31, 2024, giving ChatGPT real-time access to web information with citations.
• In February 2025, OpenAI extended ChatGPT Search to all users, including the Free tier, with no account required for some queries.
• Google’s AI Overviews reached over 100 countries and 30 languages by October 2024, including Spanish, Portuguese, Japanese, Korean, and Hindi.
• AI Overviews launched in the United States in May 2024 before extending to the United Kingdom, India, Japan, Brazil, and Indonesia within six months.

Source: Google Blog, OpenAI

For comparable platform-adoption data on AI in adjacent verticals, see our AI in social media tools coverage. For Microsoft 365’s footprint in enterprise AI workflows, see our Microsoft 365 statistics coverage.

Legal and Regulatory Pressure on AI Search

• The New York Times Company filed suit against Microsoft Corporation and OpenAI in the Southern District of New York on December 27, 2023 (case 1:23-cv-11195), alleging unauthorized use of millions of copyrighted Times articles to train large language models.
• OpenAI’s motion to dismiss was partially denied in April 2025, allowing core copyright claims to proceed in the case.
• The case remains active as of 2025 and is being closely watched as a precedent-setting test of training-data fair use.

Source: Court Listener (S.D.N.Y. docket)

For a wider context on Google’s overall search dominance and the ranking factors AI is reshaping, see our Google SEO statistics.

Frequently Asked Questions (FAQs)

Conclusion

AI now sits at both ends of the search workflow. An estimated 86% of SEO professionals have integrated AI tools, while AI Overviews appear in 13.14% of all Google searches, more than double the rate measured in January 2025, and have cut position 1 click-through from 1.41% to 0.64%. The two trends compound: SEO teams use AI to produce more content, faster, into a search environment where AI summaries absorb a growing share of clicks before users ever reach a publisher’s page.

The asymmetry across query types matters. Informational publishers face the steepest CTR damage and the largest publisher traffic declines (HuffPost down approximately 65% from peak; news publishers off 30 to 40% on Google referrals). Transactional queries hold up better, and BrightEdge data showing only approximately 20% overlap between LLM citations and Google’s top 10 suggests AI search rewards a different content profile than traditional ranking ever did. For SEO and content teams, the practical read is that this year is no longer a single-discipline market: ranking on Google, earning LLM citations, and surviving AI Overview triggers are three related but distinct optimization games, and the data above is the scoreboard.