The State of PR 2026 numbers point to the biggest opening of PR's next decade: the teams that make AI visibility measurable will own the category.

More than half of PR teams are now working to shape what AI engines say about their clients. There's one problem, though. Three out of four of those teams never check whether their work resulted in citations in the AI engines.

That finding sits inside Muck Rack's latest State of PR 2026, and it's a bigger story than the one making the rounds. 55% of PR professionals say they're securing high-authority coverage specifically to influence how their brand shows up in AI-generated answers. Yet only 25% track whether their clients' brand actually appears in those answers. And 39% measure absolutely nothing.

The industry is spending real budget to move a needle that it never measures.

Collectively, 971 PR pros answered Muck Rack, a PR software company, and more than half work at agencies. Discount for all of that and the gap still stands.

The answer is the new front door

This matters because the AI answer has become the first thing a buyer sees. When a founder, a procurement lead, the media, or an investor wants to understand a category, a growing share starts with ChatGPT, Perplexity, Claude, or Gemini instead of a traditional Google search page. The AI engine's answer is the shortlist. If your brand isn't in it, you're not losing a ranking. You're missing from the conversation that decides who gets considered for the purchase or for inclusion in the story.

So the 55% have the instinct right to secure high-authority coverage that shapes how their brand shows up in AI-generated answers. Indeed, showing up in AI answers is worth real money and real effort. The problem isn't the ambition. It's that the work stops one step short of proving it actually moved the needle.

Everyone grabbed the ownership headline

Ownership was Muck Rack's own press-release headline: who, inside the organization, owns AI visibility? It's a fair question, and the data backs the concern. Across organizations, 42% can't name who owns AI visibility at all.

But ownership is the smaller story. Assigning a name to the work is a Tuesday-afternoon decision. A CMO reads "give it an owner" and thinks: fine, I'll hand it to my SEO lead. Problem solved, and no retainer required.

The harder question isn't who owns it. It's how anyone would know whether that owner is actually moving the brand's presence in AI answers.

The accountability gap

Here's the number that turns Muck Rack's survey from interesting into uncomfortable. In the same report, 69% of PR professionals say that producing measurable results is the single best way to prove PR's value.

Sit with the contradiction. The professionals who took this Muck Rack study agree, nearly seven in ten, that measurement is how PR earns its seat. Then the field takes on what 73% of PR pros call the next frontier, AI visibility, and doesn't measure it. PR is coming up short on its own test, on the one offering where it most needs to be trusted.

That's the accountability gap. It's not a talent gap, a tooling gap, or an ownership gap. It's the gap between the work being sold and the proof that it worked. Every retainer that promises AI visibility and reports back with a traditional press clip report, instead of proof that the coverage earned citations in the AI engines, makes the gap wider.

Buyers are getting sharper too. The founder writing the check has started asking a question that PR isn't ready for: when someone asks an AI engine about my category, do I come up, and did your agency's work change that? "We landed you three placements" isn't an answer to the client's question. To the client, that answer sounds like a change of subject.

How the gap plays out

Now let's walk through it. An agency secures strong coverage in news outlets that the AI engines tend to trust. Good work, and real effort. The invoice goes out. Then nothing gets measured, because measuring would mean going back to the AI engines, running the buyer's real questions, and checking whether the brand now surfaces in answers where it didn't before.

Most PR teams are skipping this step. So, the client pays to influence AI answers and never learns whether they got what they paid for.

The reflex fix is an AI visibility dashboard. A dozen tools now track AI mentions on a live feed, and they're useful for watching. But watching isn't the job. A dashboard only shows you a live feed: whether your brand comes up when someone asks ChatGPT to name the top tools in your category, or a ping when a mention appears or drops off. That's the extent of it. A dashboard won't tell you that your company is absent because the AI engine can't find a single authoritative source that vouches for the brand, or because the basic facts about your company are inconsistent across the web. Dashboards can't write that coverage, fix the record, or earn the citation. A dashboard reports the score. It doesn't move it, because it can't.

That's the distinction the whole category keeps missing. Monitoring is a running readout, like stepping on the scale each morning. It tells you the number; it doesn't change it. Measurement is the before-and-after around work built to move the result. One watches the answer. The other changes it, then proves the change.

Closing the gap takes three moves

The fix doesn't take a new tool or a bigger budget. It takes discipline, applied in the same order every time. Three moves, and any PR team can run them.

First, get the baseline at the start of any engagement or campaign. Before any coverage goes out, run the buyer's real questions across ChatGPT, Perplexity, Claude, and Gemini and record exactly where the brand stands. This is the floor. Without a baseline, every "after" is a guess.

In one recent engagement, we baselined a client, an AI-native software company, across the four major AI engines. It appeared in only half of the questions its buyers actually asked. Before that baseline, no one could have told you the real number was 50% or 5%. And that's the whole point. You can't move what you never measured, and a quarterly re-audit is where the after shows up.

Second, do the work that changes answers. AI engines cite brands they can find and trust, which means evidence the machines can parse: authoritative third-party coverage, the same core facts about the company showing up consistently everywhere online, and primary sources they can reference. The goal is a place inside the AI engine's answer, not just a source link beneath it. It's a different job than classic SEO. Ranking a link and earning a citation are related, but they're not the same. At Ignite X, we call this discipline Machine Relations.

Third, run the re-audit. Ask the same questions of the same AI engines a quarter later. Did the brand's presence in the answers change? That quarterly re-audit isn't monitoring. It's the measurement event, the before-and-after on work built to move the answer. It's also what replaces a press clip report with a result. The re-audit is how you prove the work moved the answer, instead of asking anyone to take it on faith.

A few questions this raises

Isn't AI-visibility monitoring the same as measuring it? No. Monitoring is a live feed of mentions. Measurement is a controlled before-and-after around work designed to change the answer. One is a thermometer. The other is the treatment plus the follow-up scan.

Can't the SEO team just own this? They can own it. Owning it and measuring it, however, are different jobs, and the skills only partly overlap. SEO optimizes for ranked links. AI visibility optimizes for cited answers. Related but not identical.

What's the one metric to start with? Start with presence in buyer-intent queries: the real questions your buyers put to the engines when they're deciding what to consider. When they ask about your category, how often does your brand come up? Measure it once and you've got your starting line. Do the work that earns more mentions, run the same questions a quarter later, and see whether the number climbed. Checked twice, it's the difference between guessing and knowing your work paid off.

The service line that can't measure itself

The State of PR 2026 didn't just describe a measurement problem. It described a profession that already knows measurement is the answer yet hasn't applied it to the work that will define its next decade.

A service line that can't show its results won't survive its own value test, and 69% of the industry just said as much. The teams that make AI visibility accountable, that baseline it, move it, and measure it, will own the category. The ones still sending press clip reports will be explaining, a year from now, why they charged clients to influence AI answers without ever checking whether the brand showed up in them.

PR is selling AI visibility it can't see. The fix isn't a better dashboard or relying on traditional press clip reports. It's the discipline, in today's AI era, to measure your service and its deliverables.

Get Found. Get Cited. Get Chosen.

Sources: Muck Rack, State of PR 2026 (971 usable responses; 56% agency, 91% US).