Does social media content help your brand show up in AI search?

Do AI search engines cite social media at all?

Yes, Google’s AI Overviews cite social platforms often. YouTube, Facebook and Instagram rank among their most-cited sites.

Xu and colleagues (opens in a new tab) ran 55,393 trending US searches over 40 days in 2026 and logged every source in Google’s AI Overviews, the AI summary at the top of Google’s results. The most-cited site was youtube.com at 5.49% of all citations. Facebook followed at 3.68% and Instagram at 3.65%, both ahead of most news sites.

Our own AI Overview study of 481 US AI Overviews found the same pattern. Facebook was cited in 7.1% of them. Forums, social and video platforms together made up 20.0% of all citations. For a wider view of which kinds of sites AI engines rely on, see which websites AI search engines cite.

Which AI engines cite social platforms most?

Google’s AI features and Perplexity cite social content; ChatGPT and Claude mostly do not.

Huang and colleagues (opens in a new tab) sent 11,000 real search questions to several systems. OpenAI’s GPT-4o-mini with web search drew 0.1% of its citations from social platforms. Google’s AI Overviews drew 8.5%, and Google’s ordinary results 13.4%. Chen and colleagues (opens in a new tab) looked at searches about well-known brands. Their “social” group also counts Reddit, YouTube and Quora.

Google’s two AI features also differ. In our comparison of AI Mode and AI Overviews, on the 236 US searches where both cited sources, forums, social and video made up 18.8% of AI Overview citations against 11.2% for AI Mode, Google’s chat-style search tab.

Does a cited social post shape the answer as much as a website?

No. Google’s AI cites social posts more than it uses them, and cites them far less than ordinary results show them.

Huang’s team compared what Google’s AI summaries cited with what they actually said. Content from social media and forums was drawn on 22.1 percentage points less than other cited sources. A citation, in other words, did not mean the post shaped the answer. Our guide on whether Reddit shapes Google AI Overviews looks at the same gap for forum threads.

Google’s AI is also more selective than Google itself. In Xu’s study, 14.2% of AI Overview sources came from social and community platforms, against 49.9% of the links on Google’s first page of results.

Which industries lean most on social content?

Visual, experience-led categories lean most on social content. Institutional topics like health lean least.

Xu’s team found that the share of AI Overview citations from social and community platforms varied by a factor of 3.1 across categories. It was highest in beauty and fashion at 28.9% and autos at 27.4%. It was lowest in health at 10.7% and climate at 9.3%.

Shopping questions follow the same pattern. In an audit of AI product recommendations by Uberti-Bona Marin and colleagues (opens in a new tab), social and community sources made up 22.2% of the sources displayed in Google’s AI Overviews. Reputation questions do too. In our study of “is this brand legit?” questions, 43.0% of Google AI Mode answers cited a forum, social or video source, against 11.4% for ChatGPT.

What about LinkedIn, X and TikTok?

There is little evidence that they matter much yet. Four platforms account for nearly all social citations in Google’s AI.

In Xu’s study, YouTube, Facebook, Instagram and Reddit supplied 96.4% of the social and community citations in AI Overviews. LinkedIn, X, TikTok, Pinterest, Quora and Threads together supplied 0.51% of all citations.

LinkedIn does appear in a different dataset. Zhang and colleagues (opens in a new tab) analyzed 21,143 citations from ChatGPT, Google and Perplexity on 602 test prompts. There, linkedin.com was the fifth most-cited site, with 187 citations. That dataset was built by one of the authors and mixes English and Chinese prompts, so treat it as a hint rather than a measure. None of the studies we reviewed reported X or TikTok on their own.

What should you do about it?

Treat social content as a Google channel first and a source of real evidence, not a shortcut.

1. Check which engine your buyers use. If they rely on ChatGPT or Claude, social posts will rarely be cited, and independent reviews and press matter more.
2. If Google’s AI matters to you, keep YouTube, Facebook and Instagram profiles accurate and useful, especially in visual or consumer categories.
3. Publish things people can quote: clear product facts, demonstrations and answers to common questions, not slogans.
4. Do not seed fake reviews or posts. Research on fake reviews and fake brands in AI recommendations shows how the practice works and why it puts brands at risk.
5. Measure by engine. Track whether social sources appear in answers about your category on each AI engine before shifting budget.

If you want help deciding where social content fits, see our generative engine optimization service.

What does the research not tell us yet?

The studies count which platforms get cited. None tests whether a brand’s own posting changes its visibility.

• No study we found compared brands that post often with brands that do not, and measured their AI mentions.
• Most data covers Google in the US. Other countries and languages are thinly studied.
• “Social” groups differ between studies; some include Reddit, YouTube and Quora, which behave like forums or video libraries.
• Several studies could not read social pages behind logins, so how AI engines use their content is partly unknown.
• AI engines change often, and these figures are snapshots from 2025 and 2026.

Frequently asked questions

Does ChatGPT use Facebook or Instagram posts?

Rarely, based on current research. OpenAI’s GPT-4o-mini with web search took 0.1% of its citations from social platforms in Huang’s study of 11,000 questions.

Should we post more on Instagram to show up in AI Overviews?

It may help in visual categories, but no study has tested it. Instagram supplied 3.65% of AI Overview citations in Xu’s study, and social sources mattered most in categories like beauty and fashion.

Does LinkedIn content show up in AI search?

Sometimes, though the evidence is thin. LinkedIn was the fifth most-cited site in one dataset of 21,143 citations, but a tiny share of Google’s AI Overview citations in another.

Is YouTube more useful than other social platforms for AI search?

For Google, yes. YouTube was the single most-cited site in Google’s AI Overviews at 5.49% of citations; see our YouTube study for which videos get cited.

Sources

• Xu and colleagues (2026), Measuring Google AI Overviews: Activation, Source Quality, Claim Fidelity, and Publisher Impact (opens in a new tab), arXiv:2605.14021.
• Huang and colleagues (2026), Answer Bubbles: Information Exposure in AI-Mediated Search (opens in a new tab), arXiv:2603.16138.
• Chen and colleagues (2025), Generative Engine Optimization: How to Dominate AI Search (opens in a new tab), arXiv:2509.08919.
• Uberti-Bona Marin and colleagues (2026), "If I Had to Buy Just ONE: Galaxy S26 Ultra": Auditing AI-Generated Product Recommendations (opens in a new tab), arXiv:2609.18729.
• Zhang, He and Yao (2026), From Citation Selection to Citation Absorption: A Measurement Framework for Generative Engine Optimization Across AI Search Platforms (opens in a new tab), arXiv:2604.25707.
• Underneath (2026), Do AI Overviews cite the pages that rank? 4,051 citations
• Underneath (2026), AI Mode vs AI Overviews: how different are the sources?
• Underneath (2026), “Is this brand legit?” How AI assistants build a reputation
• Underneath (2026), The YouTube videos Google’s AI cites are small