Anna Corbett
Director of Client Success
Anna is responsible for all client delivery, and is our resident data and analytics lead.
AI search increasingly cites LinkedIn content. Anna Corbett explores why LinkedIn articles influence AI answers and how organisations can strengthen their AI visibility.
It’s no secret that users are increasingly turning to AI search platforms such as ChatGPT, Google’s AI Overviews and Perplexity to ask questions about organisations, industries and services. Instead of browsing links, they receive a single summarised answer built from multiple sources across the web.
While developing our AI Performance Analyser, we began noticing a pattern in how organisations were appearing in these AI-generated responses. LinkedIn content kept appearing.
Posts, commentary and articles for LinkedIn were frequently being referenced when AI systems generated answers about companies and professional topics. That observation has already started to shape conversations we are having with our clients.
LinkedIn has long been a platform for professional networking and thought leadership. But in an AI domain search environment it is increasingly becoming something else as well: a source that AI systems draw on when explaining industry-level expertise.
New research from Semrush, analysing 89,000 LinkedIn URLs cited in AI responses, now confirms this pattern at scale.
In this article I explain this shift and why it matters for B2B organisations. But before we start, I’ve made a short video summarising the research – and two important caveats.
As I previously shared in my article about brand visibility in AI, AI systems do not rely solely on company websites. They pull information from sources that appear credible and informative. Semrush’s analysis found LinkedIn is the second most cited domain across ChatGPT Search, Google AI Mode and Perplexity.
This should be interpreted as a strong directional signal rather than a guarantee of visibility in any individual response, but this still places LinkedIn ahead of sources such as Wikipedia, YouTube and major news publishers – second only to Reddit as AI’s most cited source. But, unlike Reddit and other publications, LinkedIn is a platform where organisations can still largely control the conversation around their brand.
For B2B organisations, the implication is simple. When buyers ask AI tools questions about an industry, LinkedIn posts and articles often shape the answer. If your organisation is not contributing to that knowledge pool, someone else will.
One reason LinkedIn is one of the most cited domains in AI tools is credibility.
Much of the content on the platform is written by named professionals with visible experience, roles and career history. That context helps AI systems assess the reliability of what is being said.
We often talk about E-E-A-T principles – demonstrating your experience, expertise, authoritativeness, and trustworthiness – in the context of good SEO, but this research shows that they apply to AI visibility too.
The Semrush research also measured how closely AI responses matched the meaning of the LinkedIn content they cited. Similarity scores ranged between 0.57 and 0.60, meaning AI responses often closely reflect the ideas and explanations contained in the original post or article.
In practical terms, this means LinkedIn content does not just appear as a citation. It often influences how AI explains the topic itself – which has huge implications if you’re asking how you can control what AI says about your organisation.
Semrush’s research also highlights clear patterns in the LinkedIn content AI systems prefer.
In short, AI systems appear to prioritise clear, useful expertise over reach or virality.
LinkedIn is no longer just a networking platform. It is becoming part of the infrastructure AI systems use to understand professional topics.
For organisations producing thought leadership, the opportunity is simple: publish expertise on your website and encourage the author to republish a LinkedIn article version under their own profile, linking back to the original piece. This strengthens expertise signals, creates additional sources AI can reference and expands the footprint of your organisation’s knowledge. While this does not guarantee inclusion in AI-generated answers, it increases the likelihood that your expertise is part of the information AI systems draw upon.
As AI search becomes a primary discovery channel, that footprint matters. The businesses that are more likely to appear in AI Search results will be those whose expertise is visible across the sources AI systems trust – not deterministically, but consistently over time as those signals accumulate.
AI platforms are already shaping how organisations are discovered and evaluated. Our AI Performance Analyser shows where your organisation appears in AI responses, how it is framed and where competitors are gaining ground.
Register your interest to receive an AI Performance Analysis when it launches in May 2026.
Director of Client Success
Anna is responsible for all client delivery, and is our resident data and analytics lead.
View my other articles and opinion pieces below
We analysed 86 UK energy organisations to understand how prepared they are for AI-driven search. Looking at the ten highest-scoring companies, five trends stand out: they describe their services clearly, give important offers dedicated space, define who those offers are for, connect them to tangible outcomes, and provide evidence to support their claims.
Will AI search mean people stop visiting your website? The honest answer is… yes, in some cases. AI-generated answers are already changing how people use search. Pew Research Center found that users clicked a traditional Google result on 8% of visits where an AI summary appeared, compared with 15% of visits where there was no […]
AI browsers are going to change how people interact with websites. Google is rolling out Gemini in Chrome to Mac and Windows desktop users in the US. This means users can ask Gemini to clarify complex information on any webpage they’re reading. Google has also introduced the ability to ask questions about the current page […]
When I was three, my brother had a stick. Not just any stick – a very good stick. I wanted it with the full commitment only a three-year-old can bring to something they have no practical use for. There were a few issues. It was too big for me, it belonged to my brother, and […]