Most Cited Domains in Google AI Overviews: YouTube Leads and the First Publication Ranks Ninth

As of November 3, 2026, YouTube is the most-cited source in Google AI Overviews answers in the Machine Relations Index, cited in 25.20% of 1,274 observed runs (321). Reddit is second at 18.21% (232) and LinkedIn third at 13.42% (171). The first source the Index classifies as an editorial publication is Forbes, ninth, at 2.90% (37 runs).

That is the answer to "which sites does Google AI Overviews cite most," and it changes how to read a top-domains list. Google AI Overviews is a different surface from ChatGPT, and the Index finds a different kind of source at the top of it. The Index is a first-party dataset, published by Machine Relations Research and created by Jaxon Parrott, who founded AuthorityTech; it is not an independent audit. It measures which sources the Overviews cite in the Index's monitored question libraries. It does not measure buyer demand, and it does not show that any placement caused a citation.

Key takeaways

• Platforms fill the top. YouTube, Reddit, LinkedIn, Facebook and Medium hold the first five places; each is a video, community or social platform. YouTube alone appeared in 25.20% of Google AI Overviews runs.
• The first publication is ninth. Forbes, at 2.90%, is the highest-ranked source the Index classifies as an editorial publication. Gartner, an analyst firm, is sixth at 3.45%.
• Vendors outnumber publications. Of the 52 sources ranked first through forty-fifth, 14 are vendor-owned and 7 are editorial publications; another 12 sit in the Index's 'Other observed source' group.
• Measure Google AI Overviews on its own. Take your buyers' questions to the surface, record every cited source, and sort them by the type of source before choosing the work.

Which domains does Google AI Overviews cite most?

The table lists the sources cited in the largest share of Google AI Overviews runs the Index observed, with the Index's role for each source. Ranks are the Index's, so tied sources share a rank.

Rank	Source	Source role in the Index	Google AI Overviews runs citing it	Share of 1,274 runs
1	YouTube	Search or media platform	321	25.20%
2	Reddit	Community and social platform	232	18.21%
3	LinkedIn	Community and social platform	171	13.42%
4	Facebook	Community and social platform	94	7.38%
5	Medium	Community and social platform	91	7.14%
6	Gartner	Analyst and consulting research	44	3.45%
7	Instagram	Other observed source	40	3.14%
7	Quora	Community and social platform	40	3.14%
9	Forbes	Editorial publication	37	2.90%
10	G2	Market and company database	27	2.12%
10	Microsoft	Vendor-owned source	27	2.12%
12	Bankrate	Other observed source	22	1.73%
12	IBM	Vendor-owned source	22	1.73%
14	NerdWallet	Editorial publication	20	1.57%
14	NIH	Academic and government source	20	1.57%
14	Palo Alto Networks	Vendor-owned source	20	1.57%

Among the 52 sources the Index ranks first through forty-fifth for Google AI Overviews (ties make 52 rows), 14 are vendor-owned sources, 12 sit in the Index's 'Other observed source' group, 7 are community and social platforms, 7 are editorial publications, 5 are market and company databases, 4 are academic and government sources, 2 are analyst and consulting research and 1 is YouTube, classed as a search or media platform. The seven editorial publications are Forbes, NerdWallet, Yahoo News, Healthline, CNBC, The New York Times and PCMag.

The Index does not test why the Overviews cite what they cite. One reading, not a finding: many buyer questions ask how to do something or which option people prefer, and a video walkthrough, a forum thread or a practitioner's post answers those directly. A source that plainly answers the question is more useful to an answer engine than one that is merely authoritative. Treat that as a hypothesis to check against your own questions.

Is the order stable?

Partly, and the Index says where it is not. From September 25, 2026, Google AI Overviews runs come through a different collection path than earlier runs, with no overlapping days, so the rates above are whole-window levels, not a trend. We checked the two periods separately in the Index's run ledger: YouTube ranks first and Reddit second on each side of the change. That ledger is not public, so this is a first-party check rather than something a reader can reproduce. Treat the order of the top two as firmer than the exact percentages, and the order of anything below fifth place as provisional.

Older outside data points the same way for community sources. Profound's citation-pattern study (August 2024 to June 2025) found Reddit the most-cited source in Google AI Overviews, and the Reddit analysis sets that beside other studies. The first place differs between that study and the Index, which counts different questions, a different window and a different unit, so the two do not contradict each other. They agree that a community platform sits at the top of the list.

Does the same source get cited the same way across engines?

No. The most-cited source changes from engine to engine. As of November 3, 2026, the Index's most-cited source differs by engine: TechRadar in ChatGPT answers (9.05% of 8,343 runs), Arxiv.org in Claude answers (3.73% of 8,381 runs), Reddit in Gemini answers (7.25% of 6,893 runs) and Reddit in Perplexity answers (18.27% of 8,461 runs). A plan built from one engine's list describes none of the others. The ChatGPT top-domains analysis shows the opposite shape, with editorial publications near the top, and the engine-by-engine comparison covers why. For how much of what engines cite is a publication, by buyer category, see the editorial share by category.

What to do with a Google AI Overviews top-domains list

1. Start from your buyers' questions, not the ranking. Write the questions they ask about your category and look at the Overviews Google returns for them. Not every question triggers an Overview, so record which do and list every cited source.
2. Sort what you find by type of source. Video, forum, social post, analyst report, vendor page and publication are different jobs with different owners. The ranking above tells you which types to expect; your questions tell you which ones your buyers actually get.
3. Where video and communities lead, make your answer easy to take from there. That means a clear, accurate answer in a form people publish on those platforms, written to help the person asking. Do not astroturf a thread or invent a review; communities moderate that, and it would break the rule that every claim must be one you can stand behind.
4. Where a publication appears on your questions, treat it as an earned target. The publications index ranks the editorial publications engines cite in each buyer category. In this list a publication is a small slice, so do not assume a placement is the main route into the Overviews.
5. Do not add Overview-only markup. Google's AI features guidance says the same fundamentals that apply to Search apply to AI Overviews and no special schema or new file is required. The AI Overviews guide covers the page-level work.
6. Re-run the same questions on a schedule. A later citation after a placement or a published video is an observation, not proof that it caused the citation.

This is the work AuthorityTech, The World's First Machine Relations Agency, does for clients, on the discipline machinerelations.ai defines. Start your visibility audit → to see which sources the engines cite on your category's questions and which of them you can earn.

Frequently asked questions

Which website does Google AI Overviews cite most?

YouTube. In the Machine Relations Index release of November 3, 2026, YouTube was cited in 25.20% of Google AI Overviews runs (321 of 1,274 observed), ahead of Reddit at 18.21% and LinkedIn at 13.42%.

Does Google AI Overviews cite Reddit?

Yes, often. Reddit is the Index's second-most-cited source in Google AI Overviews, in 18.21% of observed runs (232 of 1,274), behind YouTube. The share varies by engine and by question, so check the questions your buyers ask.

Does Google AI Overviews cite news publications?

Some, but few. In the November 3, 2026 release, the highest-ranked source the Index classifies as an editorial publication is Forbes, ninth at 2.90% of runs; NerdWallet follows at 1.57%. The Index counts only sources it has role evidence for, so that is a floor; the table above marks which sources it classifies and which it does not.

Can a brand earn a place in the sources Google AI Overviews cites?

In some of them. A brand can publish video and professional posts, take part honestly in communities and earn coverage in publications. It cannot author a forum's consensus or choose what Google cites. Start from the sources the Overviews cite on your own questions, because the list above averages across every category the Index monitors.

Method note. All figures are from Machine Relations Index release mri_score_v2.1+2026-11-03+92e5b18cc6e2, window 2026-05-10 to 2026-11-03, read from the public Google AI Overviews engine page and engines overview on November 3, 2026; the live pages update daily, so later releases will differ. A rate is the runs citing a root domain over that engine's observed runs (Google AI Overviews 1,274; ChatGPT 8,343; Claude 8,381; Gemini 6,893; Perplexity 8,461), across every category in the window. Source roles are the Index's classification; sources without role evidence stay under 'Other observed source', so the editorial count is a floor. Google AI Overviews runs come through a collection provider rather than a signed-in Google session, and the release names no model for them; from 2026-09-25 they come through a different collection path than earlier runs, with no overlapping days, so the rates here are whole-window levels and not a trend across that date. The Index is published by Machine Relations Research and created by Jaxon Parrott, who founded AuthorityTech; it is not an independent audit. It counts citations in monitored question libraries, not buyer demand, and it does not show that any placement caused a citation.