What Kind of YouTube Videos Get Cited in AI Search? February 2026 Data | Promptwatch
What Kind of YouTube Videos Get Cited in AI Search?
This report breaks down the view counts, likes, subscriber sizes, and video age of YouTube content cited by ChatGPT, Perplexity, and Google AI Overviews in February 2026. Based on 100M+ citations tracked by Promptwatch.
Klaas Foppen](/content/blog/author/klaas/index.html)·February 28, 2026
How Often Does AI Cite YouTube vs Other Social Media?
This chart shows what percentage of each AI model's citations come from social media platforms. AI engines reference YouTube's instructional videos often enough that video strategy belongs in an AI visibility plan. See also our top ChatGPT sources and top AI Overviews sources for deeper domain analysis.
4.23%
YouTube
1.68%
1.10%
0.28%
0.09%
TikTok
0.02%
X
0.01%
What this means for you
YouTube's citation value depends almost entirely on which engine you care about. Perplexity cited YouTube in 2.74% of all its citations in February and Google AI Overviews in 2.25%. On AI Overviews, YouTube even beat Reddit (1.65%) as the top social source. ChatGPT, by contrast, cited YouTube just 0.05% of the time, roughly 60x less than it cited Reddit (2.96%).
In other words, there is no single "social strategy for AI search." Google-adjacent engines and Perplexity reward video; ChatGPT overwhelmingly rewards discussion threads.
How to act on it
- If Perplexity or AI Overviews matter for your audience, put real budget into YouTube: at 2.74% and 2.25% of citations it outperforms LinkedIn, Facebook, and every other social platform except Reddit on those engines.
- Don't count on YouTube for ChatGPT visibility. At 0.05% of citations it's effectively invisible there. For ChatGPT, Reddit (2.96%) and LinkedIn (0.34%) are the social channels that actually surface.
- Match platform investment to where your buyers ask questions: check which engine your audience uses before deciding whether video or community content deserves the next content sprint.
How Many Views Do the YouTube Videos AI Cites Have?
The view counts of YouTube videos at the time AI search engines cited them. Based on Promptwatch data from Jan 31 - Feb 27, 2026.
What this means for you
Mid-tier videos win: the 10K-100K view bucket was the largest at 36% of cited videos, and nearly 80% of all citations went to videos with under 100K views. Only about 7% of cited videos had passed 1M views. As with Reddit posts, AI engines pick videos that answer the question at hand. A modest tutorial that nails a specific query beats a viral hit that grazes it.
How to act on it
- Publish videos that answer specific questions, even if the addressable audience is small. A precise "how to fix X in Y" video with 3,000 views can outperform a broad 3M-view video in AI citations.
- Optimize for retrieval, not the algorithm: put the exact question in the title, summarize the answer in the first lines of the description, and add chapters so engines can locate the relevant segment.
- Revisit your back catalog of modest performers. Videos in the 1K-100K view range are prime citation material and often just need an updated title and description to match how people phrase prompts today.
How Many Likes Do the YouTube Videos AI Cites Have?
The like counts of YouTube videos at the time AI search engines cited them. Based on Promptwatch data from Jan 31 - Feb 27, 2026.
What this means for you
Engagement signals barely register in citation selection: 43% of cited videos had fewer than 100 likes, and 78% had fewer than 1,000. Only 6 videos in the entire month's sample had 100K+ likes. Combined with the view data above, this confirms AI engines aren't using social proof as a primary filter. They're matching video content (titles, descriptions, transcripts) against the question being asked.
How to act on it
- Skip the "smash that like button" tactics if AI visibility is the goal, because likes don't gate citations. Spend that effort making the content itself more extractable: clear structure, spoken answers that stand alone, accurate captions.
- Brands without a big engaged audience can compete here: a new channel's well-targeted explainer video is citable from day one, before it has accumulated any meaningful engagement.
How Many Subscribers Do the Channels AI Cites Have?
The subscriber counts of channels whose videos AI search engines cited. Based on Promptwatch data from Jan 31 - Feb 27, 2026.
What this means for you
Small and mid-size channels dominate citations: channels with 10K-100K subscribers were the biggest bucket (34%), and about 80% of citations came from channels under 100K subscribers. Channels with 1M+ subscribers earned just 7% of citations. Niche authority beats raw audience: a focused channel that consistently covers one topic area gets picked up even without a large subscriber base.
How to act on it
- Launch a branded channel focused on your niche's recurring questions instead of paying mega-influencers for reach. A channel with a few thousand subscribers is already in the most-cited cohort.
- When you do sponsor creators, weight topical fit over audience size: a 20K-subscriber channel that owns your product category is more likely to appear in AI answers than a generalist with millions of subscribers.
- Keep channel metadata tight: the about page, video titles, and playlists should make it obvious what topics the channel is authoritative on, since that's what retrieval keys on.
How Many Total Views Do the Channels AI Cites Have?
The lifetime view counts of channels whose videos AI search engines cited. Based on Promptwatch data from Jan 31 - Feb 27, 2026.
What this means for you
Lifetime channel views tell a similar story to subscribers, with a twist: the 1M-10M total view bucket leads at 32% of cited channels, with 100K-1M at 21% and 10M-100M at 20%. Brand-new channels with under 1K lifetime views still managed 28 citations. The center of gravity is established-but-not-famous channels: enough published content to prove topical depth, well short of celebrity scale.
How to act on it
- Plan for consistency over virality: publishing steadily on one topic until your channel accumulates a real body of work is the path into the 100K-10M lifetime-view band where most citations happen.
- Interlink your videos into playlists and series around core questions in your category. Cumulative topical depth separates a cited channel from a one-hit channel with the same subscriber count.
How Fresh Are the YouTube Videos Cited in AI Search?
The age of YouTube videos at the time AI search engines cited them. Based on Promptwatch data from Jan 31 - Feb 27, 2026.
What this means for you
YouTube citations skew even older than Reddit's: 54% of cited videos were more than two years old, and nearly 72% were over a year old. Videos under a month old earned just 9 citations out of 226 (~4%).
AI engines treat YouTube as an evergreen reference library, not a news feed. The videos being cited today were mostly published in 2023-2024, which means the catalog you build now determines your AI visibility in 2027-2028.
How to act on it
- Prioritize evergreen formats (tutorials, explainers, comparisons, setup guides) over news-style content. They keep earning citations for years, while timely videos age out before AI engines ever pick them up.
- Refresh instead of replace: update the description, pinned comment, and chapters of your older high-relevance videos rather than deleting and re-uploading, so the accumulated age and watch history keep working for you.
- Set expectations accordingly: a video published this quarter is unlikely to show up in AI answers for months, so treat YouTube as a long-horizon GEO investment, not a quick win.
How we collect this data
We collect millions of prompt responses, citations, and click data from the actual user interfaces of major AI platforms: over 26 billion data points and growing. This gives us one of the largest datasets on how AI search engines cite sources and recommend brands.
Real UI monitoring
Data straight from the interfaces of ChatGPT, Gemini, Perplexity, Claude, AI Overviews, and more.
26B+ data points
Over 26 billion analyzed citations, prompts, and responses, one of the largest AI search datasets available.
Continuously updated
Refreshed constantly so the trends you see reflect the latest behavior of AI search engines.
Aggregated & public
Published freely for the GEO community, based on aggregated, non-identifiable trends.