Why Media Relations Still Needs a Human Touch in the AI Era
Nothing about working in PR for nearly two decades has felt like smooth sailing.
When I started in 2003, the media world already felt unstable. Then it started moving faster. One of my first clients asked if we “knew what this Twitter thing was all about because it might blow up.” We laughed, created an account, and got pulled into the vortex. There were no apps.
Since then, PR has been one long adaptation exercise. Newsroom layoffs. Social platforms becoming news platforms. Social platforms backing away from news. Visual apps. “Thumbstopper content,” which remains a phrase that makes my soul leave my body. Digital-first media. Influencers. Podcasts. Fake news. Independent journalists. Newsletter writers. Creators. The answer engine era. AI everywhere.
We tired.
For years, I treated the changes like weather. Watch the forecast, adjust the plan, keep moving. I joined the webinars. I read the articles. I told clients what was coming next. I took pride in staying current.
But I do not think I really stopped to process why I could keep adapting. The reason, I think, is simple: media relations never really changed at its core.
The targets changed. The tools changed. The speed changed. The inboxes got worse. But the practice stayed familiar. Have a good story. Write it clearly. Figure out who might care. Get it in front of that person in a way that respects their time. Most of all, act like a human being.
Now PR is in the middle of an AI wave, and the pressure is loud. Faster. More. Draft it. Summarize it. Audit it. Scale it. Let the machine make media relations cleaner, quicker, better.
Of course I use AI. I bounce between tools to draft, brainstorm, organize messy thoughts, and pressure-test ideas. It helps. A lot.
But after more than a year of living with AI as part of the daily corporate workflow, I keep coming back to the same belief: media relations can’t just be optimized.
Every media shift has changed the tools, not the job
If you have been in PR long enough, you have probably had at least one moment where you thought, “Well, this changes everything.”
Twitter felt that way. Suddenly, reporters were reachable in public. News broke in real time. A funny post could become a storyline before legal had finished reviewing a statement.
Then Facebook became a traffic machine, until it wasn’t. Instagram shifted the value of visual storytelling. TikTok made everyone rethink attention spans. Podcasts created new long-form interview paths. Newsletters gave journalists bigger personal platforms. Substack, Patreon, and independent publishing blurred the line between reporter, analyst, creator, and community builder.
At each turn, PR people had to learn new habits:
Where journalists spent time
How stories traveled
What counted as influence
Which formats worked
How to measure success without lying to ourselves
That last one has always been tricky.
Still, under all of it, the job stayed human. A journalist did not cover your story because your subject line hit a magic character count. They covered it because the story fit their beat, served their audience, arrived at the right time, and came from someone who seemed credible enough to answer follow-up questions quickly.
AI changes the workflow. It does not change that truth.
AI is useful, but it does not have news judgment
Let’s give AI its due. It can be incredibly helpful.
It can turn a pile of notes into a first draft. It can find holes in a messaging document. It can suggest angles you may have missed because you were too close to the subject. It can summarize coverage. It can compare pitches. It can help you get unstuck when the blank page is winning.
I have used it for all of that.
The problem starts when we confuse speed with judgment.
Media relations requires taste. Not fancy taste. News taste. The kind you build by reading constantly, sending pitches that fail, watching reporters ignore angles you loved, and learning why something matters to one audience but not another.
AI can mimic a pitch. It can produce something that looks polished. It can even produce something that sounds “PR good,” which is sometimes the problem. PR good can mean smooth, bland, and empty.
A tool can write:
“We’re excited to announce a new solution that helps brands connect with consumers in meaningful ways.”
A human should know that sentence says almost nothing.
A human can ask:
What actually happened?
Why now?
Who is affected?
What tension does this reveal?
What proof do we have?
Why would this reporter care today, not six months from now?
Are we helping this person do their job, or are we adding to their inbox sludge?
AI can assist with those questions. It cannot own them. The responsibility sits with the person sending the pitch.
The best pitches still sound like one person talking to another
When I trained people on media relations, I always came back to one rule: talk to people like a human being.
That sounds almost too simple, but it solves a lot.
It keeps you from writing a five-paragraph pitch when two clear sentences would do. It keeps you from pretending a product update is a cultural movement. It keeps you from sending a “just floating this to the top of your inbox” follow-up to someone who never asked you to float anything anywhere.
It also reminds you that the person on the other side has a real job, real deadlines, real pressures, and probably a truly cursed inbox.
The best media outreach usually has a few human qualities:
It is specific without being invasive.
It is brief without being lazy.
It shows you know the person’s work.
It gives them a reason to care.
It makes it easy to say yes, no, or not now.
It does not punish silence.
AI can help draft outreach, but it cannot replace the judgment behind those choices. It does not know that a reporter just changed beats unless you tell it. It does not know that a publication has been hit by layoffs and no longer has someone covering your client’s category. It does not know that your “friendly follow-up” might land two hours after that journalist posted about being overwhelmed.
A good PR person pays attention to context. A good PR person reads the room, even when the room is a messy mix of inboxes, feeds, newsletters, and search results.
Relationships cannot be manufactured at scale
“Relationships” can be an overused word in PR. Sometimes people use it to imply they can get coverage through personal favors, which is not how good media relations works.
A real media relationship is not transactional access. It is trust built over time.
It means a journalist knows you will not waste their time. It means you send things that fit their work. It means when you say, “I think this is worth a look,” that comment carries some weight because you have not cried wolf every Tuesday for three years.
It also means you can hear no without making it weird.
That kind of trust does not come from a database export or a prompt. It comes from restraint. From reading before pitching. From sending fewer, better notes. From admitting when you do not know something. From getting the spokesperson prepared. From following through after the story runs. From not asking for changes that are really just your client wishing the article had been an ad.
AI can help you research faster. It can help you sort contacts and scan past coverage. It can help you prepare briefing documents. But it cannot earn trust on your behalf.
If anything, AI raises the value of trust because inboxes are about to get flooded with more passable, semi-personalized outreach than ever before. When everyone can produce a decent-looking pitch in seconds, the differentiator becomes the thinking behind it.
The bar does not drop because tools get faster. The bar rises because the noise gets louder.
The answer engine era makes clarity more valuable
The rise of AI search and answer engines adds another layer to this work. People are not only finding information through traditional search results or news sites. They are asking tools for summaries, comparisons, and recommendations. Those tools pull from what they can understand, verify, and connect.
That means PR teams have to think beyond the old hit list. They need clear messaging. They need credible sources. They need consistent public information. They need earned coverage that actually says something, not just a quote stuffed into a roundup.
But this does not mean media relations becomes a content machine feeding robots.
Strong media coverage still starts with strong human decisions. What do we want to be known for? What proof do we have? Who can speak with authority? Where are the real stories hiding? What questions will people ask after reading this? What would a skeptical journalist challenge?
AI can help organize an answer strategy. It can map themes and suggest content gaps. It can show where language gets muddy. All useful.
But clear thinking has to come first.
If a company cannot explain what it does in plain English, AI will not save it. If the story is weak, automating the pitch just spreads the weakness faster. If the spokesperson is not ready, no tool can manufacture credibility in the moment.
The answer engine era rewards substance. It also exposes fluff. That should push PR people back to the basics, not away from them.
The pressure to go faster can make the work worse
PR has always had a speed problem. People want coverage now. They want a statement now. They want a trend response now. AI feeds that appetite because it makes the first draft feel instant.
Instant can be helpful in a crisis. It can be useful when a team needs raw material quickly. It can save hours on repetitive work.
But instant also creates a temptation to skip the uncomfortable part: thinking.
A fast pitch can still miss the story. A fast media list can still include people who have not covered the topic in years. A fast statement can still sound like it was assembled from spare parts in a legal waiting room.
Sometimes the slower work is the work:
Reading the journalist’s last few stories
Asking whether the angle is real
Pushing back on a weak announcement
Finding the right spokesperson
Checking whether the data supports the claim
Cutting the pitch in half
Waiting until there is actual news
That is not inefficiency. That is craft.
AI can shrink the time spent on busywork. Great. Let it. Build the initial list. Summarize the background. Draft the first version. Create the quick comparison. Pull out possible angles.
Then bring the human back in before anything leaves the building.
The future of media relations needs better humans, not fewer of them
The fear around AI often turns into a simple question: will it replace PR people?
Some tasks, yes. Some bad habits, hopefully. Nobody needs to spend half a day formatting a coverage recap if a tool can get the team 80 percent of the way there. Nobody needs to stare at a blank page out of principle.
But the work that matters most will still need people who can think, listen, edit, challenge, and connect.
The future will reward PR people who can use AI without letting it flatten their judgment. People who know when a draft is too generic. People who can tell a client, kindly but clearly, “This is not a media story yet.” People who can spot the difference between a trend and a coincidence. People who can build trust with journalists who are dealing with their own version of constant change.
That means the skill set may shift, but it does not become less human. If anything, it asks for more humanity:
Curiosity, because you cannot pitch well if you do not care how the world works.
Empathy, because there is another person on the receiving end.
Discernment, because not every announcement deserves outreach.
Clarity, because vague language wastes everyone’s time.
Courage, because sometimes the best media strategy is telling the truth internally before you ever pitch externally.
AI can support all of that. It cannot be all of that.
Use the tool, but keep your fingerprints on the work
I am not interested in pretending AI is a fad. It is here. PR teams should learn it, test it, question it, and use it where it helps.
Use it to brainstorm. Use it to sort. Use it to sharpen. Use it to get out of the weeds when your brain has turned to oatmeal after the fifth messaging doc of the day.
But do not hand it the keys to your judgment.
Before you send the pitch, read it out loud. Ask if it sounds like something you would say to an actual person. Ask if the story is clear. Ask if the recipient makes sense. Ask if you are adding value or just adding volume.
The tools will keep changing. They always have. The names will shift. The platforms will rise and fall. The dashboards will promise certainty they cannot deliver.
Media relations will keep asking for the same hard thing it has always asked for: a good story, told clearly, sent thoughtfully, by someone who remembered there is a person on the other side.
For more info: BestPRguy@gmail.com 310 920-2424 www.PublicityandMarketing.com
Comments