Most people still think of LinkedIn as a networking tool and YouTube as a place to watch tutorials and fails. Which is fine. But they’re also two of the most cited sources in AI-generated answers, and if your marketing strategy doesn’t account for that yet, you’re leaving visibility on the table.
SEMRush recently published a study analyzing 325,000 prompts across ChatGPT Search, Google AI Mode, and Perplexity. They tracked which domains AI systems actually cite in their responses. LinkedIn came in second, appearing in 11.03% of AI responses on average. YouTube landed at 8.77%. For context, both platforms outrank Wikipedia and every major news publisher in the citation rankings.
That result is worth sitting with for a second. When someone asks an AI a business question, the second most likely source it pulls from isn’t a news site, a trade publication, or even Wikipedia. It’s LinkedIn. And YouTube is right behind it.
Why These Platforms? Because AI Trusts Where Real People Share Real Knowledge
Reddit is #1, LinkedIn is #2, and YouTube is in the top five. The pattern isn’t subtle. AI systems gravitate toward user-generated platforms because that’s where content tends to be specific, opinionated, and grounded in actual experience.
Corporate blogs optimized for keywords don’t compete well with a LinkedIn post where someone explains exactly why a project went sideways and what they did about it. YouTube earns citations because video transcripts and descriptions give AI something to pull from a format people genuinely use to learn things. These platforms carry credibility because real humans with real expertise show up there consistently and say useful things.
For brands trying to show up in AI-generated answers, the implication is pretty clear: being on these platforms and publishing substantive content is no longer optional if visibility matters to you.
What the Data Says About Content That Gets Cited
SEMRush didn’t just measure citation rates. They analyzed 89,000 LinkedIn URLs specifically to understand what kinds of posts and articles AI actually pulls from. A few things stood out.
Original content wins. About 95% of cited posts were original, not reshares. Clicking repost on someone else’s article contributes almost nothing to your AI visibility.
Educational content dominates. Posts that explain how something works, share firsthand experience, or document specific results make up the majority of what gets cited. Announcements, promotions, and generic thought leadership fare much worse. AI is essentially doing what a good editor would do: surfacing the content that actually answers a question.
You don’t need to go viral. The median cited post had between 15 and 25 reactions. That’s not a lot. This isn’t about chasing engagement; it’s about publishing content that’s the clearest, most relevant answer to a question someone might ask.
Consistency carries real weight. About 75% of cited post authors had published at least five posts in the previous four weeks. And while having a larger following doesn’t hurt, accounts with smaller audiences still get cited regularly when the content is genuinely useful. You’re building presence, not fame.
For LinkedIn articles, the citation sweet spot is 500 to 2,000 words: thorough enough to answer something in depth, focused enough to stay useful throughout. For regular feed posts, the 50 to 300 word range tends to perform well.
The Double Win
Here’s what makes this an easy decision: LinkedIn and YouTube aren’t just useful for AI visibility. They’re already where a lot of your audience lives.
Every piece of content you put on these platforms is working on at least two levels. It reaches actual humans who might hire you, buy from you, or refer you. And it builds the kind of distributed presence AI systems use to decide whether your brand is worth mentioning.
Your website alone isn’t enough anymore. AI wants to see consistent signal across multiple trusted sources. When your brand shows up on LinkedIn with a useful take, on YouTube with a solid explainer, and on your own site with substantive content, you give the AI more to work with when someone asks a question that’s relevant to what you do.
One more thing from the SEMRush data: the semantic similarity between LinkedIn content and AI responses is notably high, around 0.57 to 0.60. That means when AI cites your LinkedIn content, it tends to mirror your actual message fairly closely. It’s not just a link. It’s essentially your voice showing up in the answer.
What to Actually Do
If your LinkedIn presence is mostly dormant or limited to job postings and company updates, this is a good time to change that. A consistent cadence of original, educational content from both your company page and the people who work there is the highest-leverage move you can make right now for AI visibility.
That means posts from real people with real expertise, not just corporate announcements. It means sharing what you’ve learned on a project, what a client discovered after a strategy change, or what you’d do differently if you started over. The stuff that reads like someone who actually knows what they’re talking about.
For YouTube, you don’t need a production budget. An explainer from someone on your team, a walkthrough of a case study, or a breakdown of something your clients always ask about can all contribute. The bar is usefulness, not polish.
Start with one format, commit to a publishing cadence, and don’t let perfect get in the way. The platforms are there. The audience is there. And as it turns out, the AI is paying attention too.