Here is a thought experiment: Take your university’s most cited faculty researcher, the one whose work has appeared in peer-reviewed journals, informed policy decisions and generated real-world impact. Now ask yourself: If someone typed a question about that researcher’s area of expertise into ChatGPT or Perplexity today, would your institution’s content surface?
Try it. Ask what researchers are learning about AI bias in health care, or which diets slow Alzheimer’s progression, or whether DNA ancestry tests reinforce harmful stereotypes. Chances are the answers will draw from news outlets, think tanks and Wikipedia—maybe even Reddit—long before they surface university scholarship on the subject.
We are at an inflection point in higher ed communications that most institutions haven’t fully reckoned with. Generative AI is now the first stop for a growing share of curious, intelligent people, from students and journalists to policymakers and donors, who are asking questions our researchers have spent careers answering.
The way those questions get answered in AI-mediated environments is not random. Large language models draw from the structured web: pages that clearly explain a concept, sites whose content is organized by topic and authority. Indexing, tagging and content architecture quietly determine which sources get surfaced and which get ignored.
And most of us have not done that work.
This is not a technology problem. It is an editorial priorities problem. For years, many university communications shops have treated research stories as stand-alone announcements. Publish and share once, archive forever. In an AI-mediated information environment, that approach makes our best work effectively invisible.
The Zero-Click Reality
At the CASE Editors Forum earlier this month, I sat in a fireside chat with Amy Bernstein, editor in chief of Harvard Business Review, that reframed this issue for me entirely.
She described what her team has been experiencing as AI search tools mature: Traffic to original content drops when AI surfaces a summary instead of sending the reader to the source. The zero-click answer. For commercial publications whose business depends on eyeballs, this is destabilizing to the point of devastating.
But then she drew a distinction that matters for universities.
University publishers, I argued, are not in the business of selling content. Our mandate, unlike a for-profit media outlet, is to move knowledge from the place where it is generated out into the world.
When I described our TCU Magazine research content and noted that it consistently outperforms every other category with our audiences, her response was immediate: You should absolutely be indexing for AI discoverability.
She named the framework: GEO, or generative engine optimization, the emerging counterpart to SEO in an AI-driven web, the set of practices that determines whether your content gets surfaced when an AI system assembles an answer to a relevant question.
As Search Engine Land has reported, the publishers most frequently cited in AI responses tend to be the same ones investing in structured, authoritative content systems, and the data increasingly favor institutions that treat discoverability as a deliberate strategy rather than a by-product of publication.
What GEO Actually Requires
The good news is that GEO is an extension of content strategy principles most communicators already know, applied to a new context.
It starts with the unglamorous work of consistent tagging—by topic, researcher, department and research area—that allows AI systems to recognize and draw from your expertise. It requires writing for the question being asked: research stories that bury their “so what” in the fourth paragraph are harder to surface than those that lead with the problem being solved, or that call it out in a sidebar or definitional box. And it means building thematic clusters rather than isolated pages. One story about AI bias in health care is a data point. A cluster of stories, faculty profiles and multimedia content organized around AI and health equity creates a body of knowledge AI systems can recognize as authoritative.
That last element, definitional content, is especially important. Short, clearly written explanations of key concepts are exactly what AI systems extract, and exactly what your human readers need: how deepfakes exploit First Amendment protections, what makes a Mediterranean diet neuroprotective, how mental health diversion courts differ from traditional sentencing. The goals are not in tension. The strategy is the same.
None of this requires a technology overhaul. It requires editorial intention applied systematically.
The Ethical Argument
At a moment when public trust in expertise is fragile, allowing credible research to remain invisible in AI-mediated environments is more than a communications failure. It is a civic one.
Our researchers are producing work that matters: evidence that a TCU-built police leadership training model has helped 75 percent of graduates earn promotions and is now studied as a national model, scholarship showing that only 37 percent of the 3.5 million college students with disabilities disclose to their institutions and research charting a path to change that, work on how facial recognition technology has made it nearly impossible to move through public life anonymously and why the legislation hasn’t caught up. If that work is invisible to the systems now controlling how most people encounter information, we have failed our researchers and our missions—not because we didn’t publish, but because we didn’t make it findable.
This is not a challenge specific to my institution. At the Editors Forum, communicators across higher education described the same pattern: years of rigorous coverage, carefully reported and fact-checked, effectively absent from the AI-generated answers their audiences are now receiving.
Discoverability is not a marketing concern. It is an editorial responsibility.
The question is not whether AI is reshaping the information environment. It is. The question is whether our institutions show up with the rigor and credibility they’ve earned, or whether someone else fills the void.
