TL;DR

For years, we’ve been hearing that traditional media relations are dying. Then AI showed up and did something unexpected: it made credible third-party coverage even more valuable.

AI engines need credible sources to construct answers and determine authority. But a media hit by itself isn’t the goal. The opportunity is to connect storytelling, earned media, owned content, shared media, and amplification into a system that repeatedly demonstrates expertise and credibility.

That’s Visibility Engineering™ and exactly why the PESO Model® matters even more in the AI era. 

Key Insights

• AI has given media relations a new role because earned media creates third-party credibility and validation that can influence the sources AI engines surface.
• Muck Rack found that earned media accounted for 84% of the AI citations it analyzed across ChatGPT, Claude, and Gemini.
• Getting coverage is no longer the finish line. The goal is to create credible signals that reinforce one another across the web.
• Storytelling matters because both people and machines need a clear, consistent narrative to understand what an organization knows and why it matters.
• Owned and earned media have become especially powerful together: owned establishes your expertise and earned independently validates it.

The Machines Need a Good Story, Too: Media Relations in the AI Era

I have spent a large part of my career in public relations and corporate communications. And for a large part of that time, someone has been predicting the death of media relations. Social media, brand journalism, influencers, and content marketing, you name it. They were going to kill all of it.

Then, newsrooms started shrinking, many publications disappeared, journalists were laid off, and getting meaningful coverage became harder. For a while, it looked like maybe the predictions were finally right, and then AI arrived.

And, somewhat ironically, the technology that was supposed to disrupt practically everything may have made ‘old-fashioned’ media relations more important than it has been in years. Just not for the reasons we’re used to.

Your Audience Is No Longer the Only Audience

In my experience, the logic behind media relations was pretty straightforward. You have a story, you convince a journalist that it is interesting, the journalist covers it, and their audience sees it (yes, I’m massively oversimplifying a process that has never been quite that easy).

Hopefully, that article created awareness, credibility, website traffic, leads, or some combination of those things. But there was always another benefit to having The Wall Street Journal, CNN, TechCrunch, or an important trade publication say something about you instead of you saying it about yourself: third-party validation. 

That still matters greatly in my book, but now there is another audience consuming that story: the machines. When someone asks ChatGPT, Claude, Gemini, Perplexity, or Google a question about your industry, company, or product, those systems need information to construct an answer and somewhere to get it. 

This is where things get interesting for PR and marketing pros. Muck Rack analyzed more than 25 million links cited in responses from ChatGPT, Claude, and Gemini across 17 industries. It classified 84% of those citations as earned media, with journalism specifically accounting for 27%. Paid and advertorial content? 0.3%. And this wasn’t a one-time anomaly. Across three editions of the study, earned media has consistently accounted for between 82% and 89% of citations.

After years of explaining why earned media still matters, AI is making that argument for us.

The Media Hit Has a New Job

This changes how I think about media placements. Imagine you run a cybersecurity company. You publish an excellent article explaining a new type of attack. Your researchers have original data, your CTO has a strong point of view, the content is technically sound and useful. That is valuable owned media.

But you are still the cybersecurity company saying the cybersecurity company knows cybersecurity. Now imagine a respected cybersecurity journalist interviews your CTO about the research. A major trade publication covers it. Another journalist references the findings and a podcast discusses the issue and mentions the company. Suddenly, the internet contains several independent signals connecting: your company → cybersecurity expertise → this particular problem.

A human sees credibility while an AI engine sees corroboration. That distinction matters.

Ahrefs studied 75,000 brands and found that branded web mentions had a 0.664 correlation with visibility in Google AI Overviews, compared with just 0.218 for backlinks. Correlation isn’t causation, of course, but that’s a pretty significant difference.

There’s another important wrinkle here. A mention isn’t automatically a useful mention. If your company appears in 100 articles but none of them connect you to the expertise you want to own, you’ve created visibility without much meaning. The context matters. What are people saying about you? What topics consistently appear alongside your brand? What expertise are journalists, analysts, customers, and other credible sources associating with you?

That’s where storytelling becomes more than the thing that gets the journalist interested. It helps create the associations you want both humans and machines to understand. And that sounds remarkably similar to what PR professionals have been trying to do all along: build reputation and authority, and get credible people talking about you.

Of course, none of this means we should go back to spray-and-pray pitching. The objective isn’t more mentions, but a body of credible evidence around subjects where you legitimately have expertise. 

AI Hasn’t Made Storytelling Less Important

We talk so much about optimization now that it’s easy to make communications sound like an engineering exercise: schema, entities, citations, prompts, keywords, structured content, and AI share of voice.

All of those things have a role, but nobody wants to read your schema! Journalists don’t cover companies because their executives have excellent metadata, and people certainly don’t remember a perfectly optimized corporate message because it appeared 17 times.

They remember stories. I learned this repeatedly while working in technology and cybersecurity communications. Some of the subjects are incredibly technical: malware analysis, threat intelligence, vulnerabilities, and nation-state attacks.

Our researchers could hand us pages of fascinating technical analysis about a new threat. Fascinating to them, anyway. My first question was usually some version of “so what?” Not because the research wasn’t important, but because our job was to translate what they had discovered into why a journalist, and ultimately their reader, should care. That’s what makes a journalist interested in the story, and the resulting coverage useful, memorable, and discoverable. AI has increased the number of places a good story can travel.

Welcome to the Trust Supply Chain

Gini Dietrich recently described AI visibility as a trust problem and trust as a supply chain. I think that’s spot on. You can’t publish 100 AI-optimized articles about how innovative your company is and magically manufacture authority. 

Consider how we behave as humans. If I tell you, “I’m one of the world’s leading experts on cybersecurity.” Your reaction is probably, “That’s nice, Alejandro.”

If a respected cybersecurity journalist says it, an industry analyst cites my research, a conference invites me to speak, other experts reference my work, and several credible publications quote me on the subject? Now you might actually believe it.

AI engines don’t evaluate trust exactly the way we do, of course. But they face a similar information problem at enormous scale: which sources and signals should inform the answer. That’s why Visibility Engineering™ isn’t about gaming an algorithm. It’s about creating the conditions in which credible signals reinforce one another and trust can accumulate. 

From Media Relations to Visibility Engineering

Traditional media relations often looked like this: pitch, land coverage, celebrate, add logo to website, send monthly report, move on, and try to repeat. Visibility Engineering asks, “What happens before and after the coverage?” 

Maybe you begin with a substantial piece of owned research. That research gives you something genuinely useful to pitch, and the earned coverage validates the expertise. You share the coverage through executives, employees, and social channels, and use paid to amplify the strongest content. 

AI engines encounter the same expertise across multiple trusted sources, and the next time someone asks, “Who are the experts on this issue?” you have increased the probability that your organization becomes part of the answer. 

That also changes how we evaluate media relations. Instead of stopping at “how many people saw the coverage?” we should also ask “did it strengthen the association between our organization and the expertise we want to own? Did credible sources reinforce it? Can people and AI engines find and understand it?” That’s a very different evaluation of a successful media hit. 

AI Is Making the PESO Model® More Relevant

The PESO Model wasn’t created for generative AI; it didn’t need to be. The basic premise was already right: paid, earned, shared, and owned media become significantly more powerful when they work together rather than independently.

AI has simply created another reason integration matters. And when those four work together consistently, something more valuable happens: you build authority. Authority builds trust. Trust creates visibility. And visibility increases the likelihood that both humans and machines include you in the conversation.

This is communications work. We understand audiences, reputation, relationships, and third-party credibility. And, most importantly, we understand how to find and tell stories that other people actually want to repeat.

The Best Communications Strategy Might Be an Old One

There is something wonderfully circular about all of this. We spent the past 20 years adding more technology to communications: search, social media, marketing automation, analytics, content platforms, and now AI.

The audience has evolved, the tools have changed, and the machines are listening. And after all of it, one of the most important strategies for succeeding in an AI-driven information environment turns out to be something PR professionals have been doing for more than a century: earn the trust of credible people and give them a story worth telling.

The difference is that the story doesn’t stop when the article is published anymore. It becomes part of a much larger information ecosystem. It gets searched, shared, referenced, summarized, connected to other sources, and increasingly fed into the systems people use to decide what companies they know, what experts they believe, and whom they trust. 

Media relations matter in a new way, and not because we suddenly became nostalgic for news clippings. They matter because trust has become infrastructure. Storytelling creates the idea. Earned media validates it. Visibility Engineering connects it. 

Now What?

Where teams often get this wrong is treating the media hit as the finish line. The coverage lands, but nothing connects to it. There’s no strong owned anchor for the journalist to reference, no shared amplification, and no broader record for the machines to find. One great hit, zero compounding.

If you’re sitting on strong research or expertise and the coverage never seems to compound, that’s a 30-minute conversation we’d love to have with you. Shoot us an email and let’s talk about where the system may be breaking down.

The PESO Model Diagnostic® is also a good place to start. It’s free, it’s scored, and it will show you where your communications program is building authority and where the pieces still aren’t working together.

And if you’re ready to build the system, the PESO Model Certification® will show you how to connect paid, earned, shared, and owned media so the work builds on itself instead of disappearing after the hit.

© 2026 Spin Sucks. All rights reserved. The PESO Model is a registered trademark of Spin Sucks.