Last updated: 14 November 2026
Reddit is the single most-cited domain across AI-generated answers, followed by YouTube and LinkedIn, according to an analysis of 30 million sources spanning ChatGPT, Google AI Mode, Gemini, Perplexity and AI Overviews. Wikipedia dominates specifically within ChatGPT, accounting for 47.9% of its top-10 citations — but the picture shifts sharply once you look engine by engine.
Key Takeaways
- Reddit is the most-cited domain overall across AI-generated answers, ahead of YouTube and LinkedIn, based on a 30-million-source study from Search Engine Land / Peec AI (2026).
- Wikipedia accounts for 47.9% of citations within ChatGPT's top 10 most-cited sources, while Reddit leads both Perplexity (46.5%) and Google AI Overviews, per Semrush/Decoding (2025).
- Only 11% of domains are cited by both ChatGPT and Perplexity, showing the two engines pull from largely different source pools, according to the 2025 AI Visibility Report.
- Referring domains are the single strongest predictor of ChatGPT citation likelihood, with sites over 32,000 referring domains 3.5x more likely to be cited than smaller sites, per SE Ranking (2025).
- Only 38% of AI Overviews citations come from pages ranking in Google's traditional top 10 results, meaning classic SEO rank and AI citation are two separate games, according to Digital Applied (2025).

What counts as an AI citation?
An AI citation is a direct reference — a link, source attribution or named mention — that an AI engine such as ChatGPT, Perplexity, Google AI Overviews, Gemini or Microsoft Copilot includes alongside a generated answer to show where a claim came from. It differs from a traditional Google search ranking because the engine is choosing one passage from one page to quote or paraphrase, rather than listing ten competing links for a user to click through themselves.
This distinction matters for any UK business professional trying to understand visibility in 2026. A page can rank nowhere on page one of Google and still be the exact passage an AI engine lifts into its answer — and a page ranking first can be ignored entirely. Analysis of 863,000 search engine results pages found that only 38% of citations in Google's AI Overviews come from pages ranking in the traditional top 10 organic results, according to Digital Applied (2025). Generative engine optimisation (GEO) — the practice of structuring content so AI systems select and cite it — has therefore become a distinct discipline from conventional SEO, which is the specialism Aether AI tracks across six engines for its users.
Which AI engines cite domains differently?
ChatGPT, Google AI Overviews, Perplexity, Gemini and Microsoft Copilot each draw from noticeably different source pools, and citation behaviour is not consistent across them. Wikipedia and Reddit dominate ChatGPT's US citations, each accounting for roughly 12–13% of all ChatGPT citations, according to Similarweb AI Search (2025).
Look closer at the top-10 breakdown and the gap widens further. Wikipedia makes up 47.9% of citations within ChatGPT's top 10 most-cited sources, while Reddit leads Google AI Overviews at just 2.2% and Perplexity at 6.6% within their respective top-10 sets, per Profound (2025). A separate analysis puts Reddit's share of Perplexity's top-10 citations even higher, at 46.5%, as part of a finding that the leading cited domains — Wikipedia, YouTube, Reddit, Google properties and LinkedIn — together account for 38% of all AI citations across platforms, according to Semrush, via Decoding (2025).
The practical takeaway: optimising for "AI search" as a single target is a category error. A business professional checking visibility needs engine-by-engine data, because the overlap between engines is genuinely small — only 11% of domains are cited by both ChatGPT and Perplexity, according to the 2025 AI Visibility Report. That is why Aether AI tracks citations separately across ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Copilot rather than reporting a single blended "AI visibility" score.
Engine citation comparison
| AI Engine | Dominant cited domain type | Notable figure |
|---|---|---|
| ChatGPT | Wikipedia, Reddit | Wikipedia = 47.9% of top-10 citations (Profound, 2025) |
| Perplexity | Reddit = 46.5% of top-10 citations (Semrush/Decoding, 2025) | |
| Google AI Overviews | Reddit (low share) | Reddit = 2.2% of top-10 citations (Profound, 2025) |
| Cross-engine (all five) | Reddit, YouTube, LinkedIn | Top of 30m-source ranking (Search Engine Land/Peec AI, 2026) |
What types of domains do AI engines cite most often?
AI engines overwhelmingly favour a small set of domain types: user-generated-content platforms (Reddit), video platforms (YouTube), professional networks (LinkedIn), and reference encyclopaedias (Wikipedia). These four categories, plus Google's own properties, collectively account for 38% of all AI citations, according to Semrush, via Decoding (2025).
This is a marked shift from the government and academic sources that dominate traditional SEO authority-building advice. Structured data, forum discussions and crowd-sourced Q&A content appear to carry weight precisely because they contain direct, specific answers in conversational language — the format large language models are trained to extract. Institutional domains do still feature, but not as heavily as the "trusted zone" assumption suggests: SE Ranking's analysis of 129,000 domains found .gov and .edu domains averaged 3.2 citations compared to 4.0 for sites without any trusted zone designation, per Search Engine Journal/SE Ranking (2025).
News publishers, niche industry blogs and company websites make up the long tail beneath these giants. For a UK business, this means competing head-on with Wikipedia or Reddit for a broad query is close to futile — the more realistic opportunity is being cited for a specific, narrow, well-defined question where no Reddit thread or Wikipedia page exists.
How do AI engines decide which sources to cite?
AI engines select citations primarily based on backlink profile strength, content specificity and topical trust signals rather than domain age or brand size alone. The single strongest predictor identified so far is referring domains: sites with over 32,000 referring domains are 3.5 times more likely to be cited by ChatGPT than sites with up to 200 referring domains, according to SE Ranking (2025).
As SE Ranking put it in its own research summary: "What ultimately matters is not the domain name itself, but the quality of the content and the value it provides." That finding is echoed in how Aether AI's content and citation-tracking systems are built — the platform treats specificity, not domain reputation, as the controllable lever.
"Engines cite pages that answer a question completely in one place, say something specific, and come from a source that looks like it should know. Vague pages summarising other pages don't get cited — they get paraphrased without credit. Specificity is the whole game." — Lauren Dawkins, Head of Content, Aether AI
Freshness, structured data (schema markup) and clean heading hierarchy also play a measurable role: a clean H1/H2/H3 hierarchy, rather than bold text used as pseudo-headings, helps retrieval systems identify where one passage ends and the next begins.
How can a UK business check if its site is being cited by AI engines?
A business can check AI citation status by manually prompting each engine with its target queries, or by using automated tracking software that scans multiple engines on a schedule. Manual checking — typing a handful of questions into ChatGPT, Perplexity, Copilot, Gemini and Google AI Overviews and noting whether your domain appears — is free but unscalable and gives no historical trend.
Automated citation tracking solves the scale problem. Aether AI's platform runs citation tracking across six AI engines — ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot — alongside keyword and competitor tracking and Google Search Console integration, so a business can see not just whether it's cited, but how that compares to named competitors over time. Aether AI's free AI-visibility audit, available at /audit, gives any UK business a starting baseline before committing to ongoing tracking.
That baseline matters because visibility is often lower than businesses assume. Aether AI's own measured baseline — published at aether-ai.co.uk/insights/ai-visibility-baseline-2026 — ran 700 live AI engine scans of UK brands' target keywords between July and August 2026, and found the brands were named in just 23 answers (3.3%); the longest-established brand, in security services, was named in 6.3% of its 270 scans. That is consistent with how concentrated the top-cited domain list is: most businesses are competing for a small residual share once Reddit, Wikipedia, YouTube and LinkedIn have taken the lion's share of citations.
A related gap shows up in Google Search Console itself. On two properties connected to Aether AI's founding team — Priority First and Aether Agency — the share of GSC clicks from queries Google does not name reached 82.1% and 88.3% respectively between 19 July and 15 August 2026. Aether AI is built and operated by Aether Agency Ltd. Priority First, Pulse Operations and Aether Agency are businesses connected to the founding team. That hidden-query share is a strong proxy for how much AI-driven and zero-click search activity is already happening beneath standard reporting.
Does citation pattern volatility matter for UK businesses?
Citation patterns are not static — AI engines reshuffle which domains they cite over weeks and months, and some events have caused sudden, significant drops for previously dominant sources. A widely-discussed volatility event in September 2025 saw ChatGPT's citation of Reddit and Wikipedia abruptly drop, though the precise causes remain debated across the industry.
"The pattern that repeats: answers move constantly, sources move slowly. Engines redraft their words week to week, but the underlying set of cited domains is sticky. Winning a citation is hard; once you're in the source set, you tend to stay — which is why starting early compounds." — Lauren Dawkins, Head of Content, Aether AI
This sticky-but-shifting dynamic is precisely why one-off audits lose value quickly. A citation snapshot taken in spring 2026 may bear little resemblance to the picture six months later, particularly as Google AI Overviews, Perplexity and ChatGPT each update their retrieval and ranking models independently. Continuous tracking, rather than a single audit, is the only way to know whether a domain's citation share is rising, falling or holding steady.
Your AI citation checklist
- Identify which of the six major engines (ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini, Copilot) matter most for your customer queries.
- Run your core business questions manually through each engine and log whether your domain is cited.
- Build out referring domains through genuine coverage and partnerships — referring domain count is the strongest predictor of ChatGPT citation.
- Structure every page with a clean H1/H2/H3 hierarchy and define key terms in plain language on first use.
- Write answers that are specific and complete in one place, rather than vague summaries that invite paraphrasing without credit.
- Add structured data (schema markup) to key pages to help retrieval systems parse content accurately.
- Monitor citation share monthly rather than once, since patterns shift and sudden drops do occur.
- Use a free baseline audit, such as Aether AI's at /audit, before investing in a full tracking programme.
FAQ
Which domains do AI engines cite most?
Reddit is the most-cited domain overall across AI-generated answers, followed by YouTube and LinkedIn, based on a 30-million-source study covering ChatGPT, Google AI Mode, Gemini, Perplexity and AI Overviews, according to Search Engine Land/Peec AI (2026). Wikipedia separately dominates ChatGPT specifically, at 47.9% of its top-10 citations.
Why does Reddit rank so highly across AI engines?
Reddit's forum format provides direct, conversational, specific answers to narrow questions, which matches how large language models extract citable claims. Reddit leads Perplexity's top-10 citations at 46.5% and tops the cross-engine ranking of 30 million sources, according to Semrush/Decoding (2025) and Search Engine Land/Peec AI (2026).
Do .gov and .edu domains get cited more by AI engines?
Not necessarily. SE Ranking's analysis of 129,000 domains found .gov and .edu domains averaged 3.2 citations compared to 4.0 for sites without any trusted zone designation, according to Search Engine Journal/SE Ranking (2025). Referring domain strength mattered more than the top-level domain type.
Does ranking well on Google guarantee an AI citation?
No. Only 38% of citations in Google's AI Overviews come from pages ranking in the traditional top 10 organic results, based on an analysis of 863,000 search engine results pages, according to Digital Applied (2025). AI citation and organic ranking are related but distinct outcomes.
How much overlap is there between ChatGPT and Perplexity citations?
Very little. Only 11% of domains are cited by both ChatGPT and Perplexity, according to the 2025 AI Visibility Report, indicating the two engines retrieve and select source material through substantially different methods.
How can a UK business track its own AI citation performance?
A business can test manually by prompting each engine with target questions, or use automated tracking across multiple engines for ongoing monitoring. Aether AI offers citation tracking across six AI engines plus a free starting-point audit at /audit.
How often do AI citation rankings change?
Citation patterns shift over weeks and months rather than staying fixed, with at least one documented sharp drop in Reddit and Wikipedia citations within ChatGPT occurring in September 2025. This means a single audit becomes outdated relatively quickly and ongoing monitoring is more reliable than a one-off check.
Tracking your citation visibility with Aether AI
Understanding which domains dominate AI citations is only useful if a business can see its own position within that landscape, and that is precisely the gap Aether AI's platform is built to close. Rather than guessing whether ChatGPT or Perplexity mentions your brand, Aether AI runs structured citation tracking across ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot, alongside keyword and competitor tracking and Google Search Console integration.
Aether AI's own measured baseline — 700 live AI engine scans across UK brands' target keywords in July and August 2026 — found an average citation rate of 3.3%, rising to 6.3% for its longest-established brand. That figure gives any business a realistic benchmark rather than a vague promise, and it reflects the same dogfooding approach the platform applies to its own visibility.
If you want to know where your business currently stands among the domains AI engines actually cite, start with Aether AI's free AI-visibility audit at /audit, or review the public pricing at /pricing to see what ongoing tracking involves.
