Does YouTube help you get cited in AI search answers?

Which AI platforms actually cite YouTube videos?

Perplexity and Google's two AI surfaces do almost all of it. In Otterly.ai's study, Perplexity drove 38.7 percent of YouTube citations, Google AI Overviews 36.6 percent, Google AI Mode 19.6 percent, and ChatGPT 4.4 percent.

Otterly.ai's YouTube citation study, published March 2, 2026, counted more than 100 million AI citations across ChatGPT, Google AI Overviews, Google AI Mode, Perplexity, Microsoft Copilot, and Gemini over a 30-day window. Social media made up 5.54 percent of everything cited. Inside that slice YouTube took 31.8 percent, second to Reddit at 46.4 percent. Copilot accounted for 0.5 percent of YouTube citations and Gemini for 0.2.

Google's side of the picture holds up in a second dataset. Ahrefs' ranking of AI Overview citations put youtube.com first at 22.9 percent mention share, meaning its slice of the citations the top sources earn. Our piece on whether ChatGPT and AI Overviews cite the same sites looks at how far those engines drift apart.

What none of this explains is why ChatGPT passes over YouTube. In ChatGPT and Perplexity alike, Reddit was the most cited social domain, at 53.6 to 62.8 percent of social citations, yet Perplexity still cites YouTube heavily and ChatGPT barely does. We have not found a source that says why, and we are not going to invent one.

Does a video's view count affect whether AI cites it?

Barely, by Otterly.ai's numbers. View count, like count, and subscriber count all showed a near-zero correlation with citation frequency, with r between -0.02 and -0.03.

In this dataset, knowing how many people watched a video told you almost nothing about whether an AI tool cited it. Small channels were well represented, too. Of the cited channels, 35 percent had fewer than 10,000 subscribers, and the median cited channel had 41 videos.

The strongest signal Otterly.ai found on a single video was the length of its description, at r 0.31. That is a correlation and nothing more, and whether writing longer descriptions would change anyone's odds is a question the study leaves open. Our piece on domain authority and AI citations asks the same popularity question about websites.

Do long videos get cited more than Shorts?

Yes, by a wide margin. Long-form videos made up 94 percent of YouTube citations in Otterly.ai's study, and Shorts made up 5.7 percent.

The largest duration band was 10 to 20 minutes, at 32.1 percent of citations. That is the biggest single band and still under a third of the total, so no one length owns the result.

We read the Shorts number as a question of room. A 40-second clip has very little space to answer a buyer's question all the way through, and a 15-minute walkthrough of the same question has plenty, with a description underneath it that can run as long as the answer needs.

Should you make video content specifically to get cited, or for people first?

For people first, with one specific question per video. Nothing in this data says a citation follows from publishing, so we would only make a video your buyers would choose to watch anyway.

Start with where your buyers ask. If they mostly ask ChatGPT, YouTube is a weak route on these numbers. If they ask Google or Perplexity, a long video that answers a question they actually type has a real shot at showing up among the sources. Our AI search guide covers the groundwork that comes before any single channel.

Alongside Corvan AI, we run the AI search work for clients who want a team on it. The median cited channel in Otterly.ai's data had 41 videos. To us that reads like a channel that kept answering questions for a while, and the study does not say how long that while was.