AI search engines do not mostly cite social networks, despite the attention social content gets in AI visibility advice. Vilnux Research tested the idea on 18,902 answers with sources about 36 Spanish brands, measured between October 12 and November 15, 2026, and found that only 9.5% of the answers cited any social network at all. Among the ten most cited third-party domains, there was exactly one social network: LinkedIn, in third place.
The study classified every cited URL by social network and piece type, counting only answers that had at least one source, and leaving out the measured brand's own site. Across 192,085 citations to websites other than the brand's own, LinkedIn took 1,976 citations and appeared in 5.7% of answers, ranking third among all third-party domains. YouTube took 815 citations and appeared in 2.7% of answers, ranking thirteenth. Facebook managed only 63 citations, appearing in 0.3% of answers and ranking 436th, and X took 57 citations, also 0.3%, ranking 514th.
The striking detail is how engine-specific each network is. Of the YouTube citations, 95.7% came from Google AI Mode, and 82.5% of the LinkedIn citations came from Perplexity. Facebook's few citations came mostly from Google AI Mode too, at 89%, while X split across engines with Perplexity supplying about half. Each network depends on one engine, which means a social presence that earns citations in one engine may be invisible in the others.
The YouTube finding has a practical twist: 61.2% of citations to YouTube videos take the reader to the exact minute of the video. That means AI Mode is not citing videos as a vague whole; it is pointing at a specific moment inside them. For video creators, the implication is that a long video with one strong minute can earn a citation for that minute, and that chapters, clear speech and a transcript that names the topic at the right point all help the engine find it.
Vilnux is careful about the limits. The brands are Spanish, in Spanish, from professional services, industry, real estate, health and education. In consumer goods or in English, social networks may weigh far more. The study also measures what AI links as a source, not what it reads: an engine may use a video or a thread without citing it. And not every brand was measured with the same engines or the same weeks, so the split across engines partly reflects how each brand was measured.
The headline number should cool the idea that posting on social networks is a shortcut to AI visibility. Nine out of ten answers cite no social network at all, and Facebook and X barely register in this panel. The two exceptions are worth the effort, though, and they are different jobs. LinkedIn is the citation source for professional and business-to-business questions, and it is Perplexity that cites it, so a LinkedIn presence that publishes substantive posts and articles is really a Perplexity visibility play. YouTube is the citation source for Google AI Mode, and it rewards videos that answer a question at a findable minute.
For link builders, the study also reframes what a good placement is. A link from a LinkedIn article or a YouTube description is not the same as a citation of that content by an AI engine, but the two are related: the content that gets cited is the content that answers a question well, wherever it lives. The practical move is to treat LinkedIn and YouTube as citation surfaces in their own right, with the same care given to a publisher placement, rather than as distribution channels for links.
Watch whether the pattern holds in English-language panels and in consumer markets, where social content may carry more weight. Watch whether Perplexity keeps leaning on LinkedIn as its professional source, and whether Google AI Mode extends its minute-level YouTube citations to other video platforms. And check your own AI visibility reports: if LinkedIn or YouTube appears among your cited sources, note which engine cited it, because that tells you which audience the content is actually reaching.
Sources: Vilnux Research, November 16, 2026.
Related reading on this site: AI Overview Citations: What 15 Studies Agree On, and Don't, Reddit Takes 29.9% of Gemini Citations in Ahrefs\' November Ranking, Wikipedia Leads ChatGPT Citations at 10.1% in Ahrefs\' November Ranking, Only 2.9 Percent of AI Citations Point at the Brand\'s Own Domain, Perplexity's Citation Funnel: 29 Query Variants, 10 Pages Retrieved, 3 Cited, and AI Chatbots Sent Shoppers to Fake Stores in Up to 15% of Answers, Study Finds.
Linkbaits helps you create content that earns citations and backlinks. Start with the toolkit and build your first linkable asset this week.
Start free trial