I spent years pitching journalists, building media relationships and celebrating every great piece of coverage we secured. And believe me, getting a brand featured in a top publication was always a reason to celebrate!
But lately, I have been wondering whether getting that coverage is still enough.
A while ago, I started learning about GEO (Generative Engine Optimisation), mostly out of curiosity about how AI is changing the way people discover brands. And then I started thinking about PR.
What happens to all those articles we work so hard to secure? Do they influence what ChatGPT or Claude says about our brands? And if they do, how would we even know?
The PR playbook that worked just a few months ago might already need a few extra pages.
The basis has not changed, but the way people discover brands is evolving fast. So what does this mean for media relations? And how do we adapt?
Earned media has a new audience
Think about the questions people are starting to ask AI assistants.
“What is the best way to get around Barcelona without owning a car?”
“Which skincare brands are actually worth the money?”
“Where can I buy good-quality second-hand furniture online?”
AI assistants might pull information from news articles, reviews, company websites and many other sources to put together their answers.
And the person asking might never visit any of them.
For those of us working in communications, this is quite a shift.
Our media coverage could influence how people discover and evaluate brands, even when they never read the original articles.
Of course, one thing I have learned is that not every article influences AI recommendations, and appearing as a source does not automatically make a brand more likely to be recommended.
But it certainly makes me wonder whether we should be looking at media relations a little differently.
PR strategies are already changing
According to Muck Rack's 2026 State of PR report, 55% of PR professionals surveyed say they are securing coverage in high-authority publications to improve their visibility in AI-generated answers.
Around half are also creating more authoritative, data-driven content and optimising content for search.
I found these numbers interesting.
PR professionals are already adapting their strategies, even though we are still figuring out how much earned media actually influences AI-generated recommendations.
We all know that original research, expert commentary and useful information have always been valuable to journalists. They give them something worth writing about while helping brands build credibility and establish themselves as trusted voices in their industries.
I have spent years working with research, trends reports and expert insights to develop stories that journalists would genuinely want to cover.
But now there is another reason to think about the information we put out there: AI systems might use it too.
And that brings me to another question.
If we start choosing publications partly because we want AI systems to find and reference our content, how do we know which ones matter?
Traditionally, I have ranked media outlets based on their reputation, domain authority, reach, editorial quality and relevance to our audiences.
Should we now be looking at how frequently AI assistants mention them?
And if we do, are we at risk of chasing AI visibility instead of focusing on the people we actually want to reach? Should we even consider building a separate media list specifically for AI visibility?
Who owns AI visibility?
Something else surprised me while learning about this topic.
Companies are increasingly recognising that AI visibility matters (just look at some of the latest communications job descriptions!), but there seems to be much less clarity about who should be responsible for it or how to measure it.
And I can understand why.
Is it a PR responsibility? SEO (or now GEO)? Brand? Content? Digital marketing?
Probably a bit of everything... Just like building a brand has always been.
SEO teams understand how to make information easier to find. Brand teams work on positioning and messaging. Content teams develop information people are looking for. And PR teams help build credibility through media coverage, relationships and thought leadership.
All these functions already work together to shape how people discover, understand and experience a brand. That has not changed. What is changing is that AI-generated answers are becoming another place where those efforts need to come together.
I do not necessarily think PR should own AI visibility completely. But I definitely think we should be part of the conversation, and that we have an important role to play. It could also be another opportunity to demonstrate the impact of our work.
Because it is not just about whether a brand appears.
It is also about what the AI says about that brand, how it describes it, which sources it uses and whether the information is accurate.
And surely, that is something we, as communications professionals, should care about.
Are we measuring the right things?
This is probably the part that I found most interesting.
Throughout my career, I have measured PR effectiveness in different ways: coverage quality, publication tiers, share of voice, proactive vs. organic coverage, brand awareness and more.
I have always believed that counting media mentions alone tells us very little. A feature in the right publication, reaching the right audience, can be far more valuable than dozens of mentions elsewhere.
But AI discovery adds another layer.
How often does our brand appear when someone asks a relevant question? Is it recommended alongside competitors? What does the AI say about it? Which sources does it reference?
And perhaps the hardest question:
Can we actually demonstrate that our PR work influenced those answers?
Because measuring whether a brand appears in an AI-generated response is one thing. Understanding why it appears is another.
Different AI assistants can give different answers. Even the same assistant might respond differently depending on how a question is phrased or when it is asked.
So how do we measure progress? How do we know whether our media coverage is making a difference? And how do we separate the impact of PR from everything else influencing how a brand is perceived?
I don't have the answers yet. But I am convinced that these are questions we should all be asking.
Maybe we do not need to reinvent PR. Just rethink how we look at it.
The more I read about AI discovery, the more I realise that many of the things we have always valued in PR still matter, a lot.
Good relationships with journalists. Credible sources. Original research. Interesting stories. Useful information.
Good PR has never been just about getting a brand mentioned. It has always been about building credibility and giving people a reason to trust you.
I do not think AI changes that.
What it might change is how our stories reach people, how they influence what people learn about brands and, ultimately, how we measure the value of our work.
And perhaps that is where we should start.
Not by chasing every new GEO trend or trying to find a magic formula to appear in AI-generated answers.
But by understanding how AI systems use information, what role earned media plays and how we can measure its influence.
I am still learning about all of this, and I find it fascinating.
If AI is changing how people discover and evaluate brands, should we also be changing how we plan and measure PR?
I would love to hear what other communications professionals think.
Are you already looking at AI visibility when planning your media relations strategy? Have you started measuring it? Or, like me, are you still trying to figure out what it all means?