A student opens ChatGPT and types: “Which university would you recommend for studying biotechnology in Spain?” The AI responds. Your institution appears. It names you precisely, describes the profile of your program, and places you among the strongest options in the country.

But there’s no link. No source cited. Just the name.

Is that good or bad?

The honest answer is: it depends. It depends on what you’re chasing and how you’re building your digital presence inside the AI ecosystem. Because there are two ways to appear in a generative answer, and each one carries a different strategic value.

The two ways to appear in AI

When a tool like ChatGPT, Perplexity, or Google AI Overviews talks about your university, it can do so in two different ways.

The first is the AI Brand Mention: the AI names your institution in an answer without linking to any source or citing any document. Your brand appears in the conversation, becomes part of the response, stays in the user’s mind. But the trail is implicit.

The second is the AI Citation: the AI explicitly cites your content as the source for what it’s saying. A link appears, a direct reference, a visible anchor to your domain or to a media outlet that has covered you.

These two forms of presence are not mutually exclusive. They are also not equivalent. They have different natures, occur in different contexts, and require different strategies. Treating them as if they were the same is one of the most common mistakes among university marketing teams starting to work on AI visibility.

The GEO Playbook for Universities places them inside Pillar 2 (Authority) as two complementary levers. This article goes deeper into each one.

The specific value of each

AI Brand Mentions: being in the conversation

A mention without a link has real value, even if it’s hard to track. When the AI names your university naturally and positively, it anchors your brand in the user’s imagination. It builds familiarity. It generates presence at the moment someone is making a decision or forming an opinion.

Think of it as PR in a generative format. The student receiving that ChatGPT answer doesn’t see an ad. They see an implicit recommendation from a source they perceive as neutral. That carries weight.

AI Brand Mentions for universities are particularly valuable in the upper stages of the funnel: when the prospect doesn’t yet know what they want and is asking the AI to help them get oriented.

AI Citations: being the reference source

Citations go further. When the AI links to your content or explicitly mentions your domain as the backing for what it’s saying, it doesn’t just increase your visibility: it increases your authority.

This is the equivalent of the backlink in the AI era. Just as a quality link in SEO transferred domain-to-domain authority, a citation from Perplexity or ChatGPT transfers credibility from the AI to your institution. It tells the user: “I’m asserting this because I found it here.”

The numbers confirm it. Brands are 6.5 times more likely to be cited through third-party sources than from their own domain (Profound, 2025). This has a direct implication for your strategy: building authority in AI is not just about optimizing your website. It’s about building an ecosystem of references that points back to you.

Which platforms generate each type

Not all generative tools work the same way. Each one has source biases, distinct citation patterns, and its own audiences. Understanding these differences lets you allocate effort where it has the highest return.

Google AI Overviews generates mostly mentions. Its responses tend to pull information from YouTube, Reddit, and topical forums. Explicit citations exist, but the model favors synthesis over direct referencing. If you want to appear here, video content and presence in online communities are key levers.

ChatGPT has mixed behavior. In its chat responses, citations are scarce — the model synthesizes from training, not real-time retrieval, except when web search is enabled. When it does cite, the most frequent source remains Wikipedia, which represents around 47.9% of verified citations (Toloka AI, 2024). The strategy here points more toward building presence in the sources ChatGPT already prioritizes.

Perplexity is the engine where citations are most frequent and most visible. Its architecture is RAG-native: it retrieves, synthesizes, and links. It favors news sites with high authority, experts with a solid web presence, and regional or niche sources relevant to the query. A university with structured, up-to-date, and signed content has more chances to be cited here than on any other platform.

The practical consequence: an AI visibility strategy that only optimizes for one platform is an incomplete strategy. The multiplatform citation secret lies in building signals that work across all three environments at once.

How to earn AI Brand Mentions

Mentions happen when the AI “knows” your brand with enough depth and context to include it naturally in its answers. That means presence in the sources models consume during training and in those they retrieve during inference.

YouTube is one of the channels with the heaviest weight in Google AI Overviews and a frequent source of training data. Videos of deans explaining a program’s approach, talks by researchers, or well-structured virtual tours are directly useful assets.

LinkedIn amplifies personal authority. Articles signed by department directors, researchers, or the rector, with substantive content on the institution’s area of expertise, build presence signals that models associate with your brand.

Wikipedia remains one of the most consulted sources in LLM training. Keeping your university’s page up to date, along with those of your most relevant researchers, is one of the highest-return-per-effort actions in the entire GEO ecosystem.

PR aimed at digital media with coverage in reference publications generates the kind of mentions that models find consistently across multiple sources. The more coherence among them, the higher the model’s confidence in your brand as an entity.

How to earn AI Citations

Citations require a different kind of content: structured, authoritative, and directly useful for the model when answering a specific question.

The “Answer First” rule is the most effective principle. Content that begins with a direct answer to the user’s question — in the first 50 words of each section — is what RAG systems extract most easily. If your page about the Sustainability master’s starts with institutional marketing instead of answering what makes that program different, you’re losing citation opportunities.

FAQs with FAQPage schema are one of the most effective formats for earning citations. A well-built FAQ section, on admissions, campus life, or research methodology, combined with the right markup, lets Google AI Overviews and Perplexity extract answers directly from your content.

Signed expert authorship applies the E-E-A-T principles that AI systems inherit from Google’s quality criteria. An article signed by a researcher with a verifiable profile, clear institutional affiliation, and academic references has far more chances of being cited than the same content unsigned.

Academic databases and institutional repositories are sources Perplexity and other engines actively consult. Publishing research data in open access, with correct metadata and DOI when possible, significantly expands your institution’s citability perimeter.

High-authority media remains the most powerful citation channel. An article in an educational outlet or national reference publication that cites your research or your program carries more weight for the AI than the same content published directly on your site.

The two presences reinforce each other

AI Brand Mentions and AI Citations don’t compete with each other. They act at different moments in the user’s decision process and build different kinds of value for your institution.

Mentions build familiarity and brand expansion. Citations build authority and verifiable trust. Together, they define how present and how credible your university is in the generative ecosystem.

The good news: many of the actions that generate citations also increase mentions. And an institution that appears consistently in AI conversations, both named and cited, builds a position that is very hard to displace.

If you’re starting out, the best entry point is the GEO Playbook for Universities: the complete framework with the five pillars that structure the strategy from content to measurement. And if you want broader context, what the SEO-to-GEO evolution means for universities explains why this shift is not a passing trend.