Skip to content
The temptation to rely on generative AI for PR pitches is enticing, but using it without human intervention stands out to reporters and will likely have the opposite effect
Close to half (44%) of reporters surveyed have a negative perception of PR content generated by AI.
That finding stems from a new survey of 1,500 of them by Global Communications Results (GRC), a Santa Ana-based PR firm. A majority of respondents, 63%, claim to have 20 years of experience as a reporter, or more.
About 12% of those polled have a positive view, 38% have a neutral view, and about 6% have another unspecified view.
Among reporters who had a negative view, the objections centered around the mediocre quality of generative AI copy. Verbatim comments published in the report include the following:
- “Too promotional and lacks perspective.”
- “Reads like a bot wrote it.”
- “Cookie-cutter and not tailored to my audience.”
GRC sums up the findings this way:
“While some acknowledged AI as a tool for background data collection, the majority stressed that editorial integrity requires human insight. AI-generated material was often equated with laziness, undermining credibility for both PR professionals and their clients.”
Generative AI copy is, by definition, an average. It ingests the best copy, the worst copy, and all the copy in between, and uses the ‘average’ to produce a probable response. If you rely on probable copy, then your copy will be no different from anyone else’s.
That’s a problem because being different is a key factor in standing out in a congested marketplace of ideas. And it’s notwithstanding the proven tendency for AI to get facts wrong, “with confidence” as much as 60% of the time.
The solution is straightforward, according to the report:
“PR professionals can leverage AI for efficiency, but human editorial oversight is non-negotiable. Journalists value context, quotes, and unique insights that cannot be automated.”
(click image for higher resolution)
PR relationships with editorial
In my experience, PR and comms professionals tend to overstate the strength of their relationships with reporters and journalists. A reporter will be interested in a good story, regardless of a relationship, because it’s a good story.
It’s far easier to alienate a reporter than it is to build a relationship with one. To that end, this study underscores the notion that pitching factually inaccurate data can absolutely earn PR a reputation for being unreliable.
Among the other key findings in the study were:
1. Reporters rely on PR at least some of the time
The survey found more than 80% of reporters “rely on PR professionals for some portion of their reporting.”
In my experience, some are open to PR pitches because good PR people can be a source of scoops. There are others who downplay that reliance or are quite vocal about rejecting PR pitches – it’s a badge of honor for a few.
One thing that’s clear from the study is this: trying to pass off AI slop as a genuine PR pitch is a surefire way to achieve the latter.
2. Reporters prefer email pitches
The report says 94% of reporters prefer email pitches. Even so, “many note inbox overload and generic outreach as a pain point.”
Multiple studies I’ve looked at draw similar conclusions. Reporters can get hundreds of pitches a day. Generative AI is poised to add noise to those inboxes, which still compete for attention against genuine pitches.
Some have said the tenability of email has been on thin ice for years – yet it’s never broken to date. The savvy PR professional, or the PR software maker, who figures out a better way and gains traction, would be a big disruption.
Many have tried unsuccessfully, yet the email deluge is a constant source of motivation to continue pursuing an alternative.
(click image for higher resolution)
3. More than half rely on press releases
Another finding from the survey was that 51% of reporters cited press releases as the preferred format for PR content. I’ve seen other studies from press release distribution companies that say the same, perhaps more compellingly too, but I question the conflict of interest on those.
Yet in the age of AI, press releases do strike me as a more reliable source of information than other PR materials. This is because these are vetted documents, often approved by a committee, and represent the official viewpoint of a business.
You do not need a press release distribution service to use a press release, but there is some validity to using a service:
First, these services do perform some level of validation before approving a release for distribution. Second, once the release is distributed, it cannot be changed or edited on syndicated sites – it’s a permanent record of sorts.
By contrast, a business can edit a blog post, and no one may notice.
Stop hiring comms people who talk and write like corporate robots. They will lead you astray.
Hire comms people who are culturally attuned, slightly unhinged, and witty, they will lead you to the promised land.
— Cristin Culver (@CristinCulver) October 24, 2025
Original thought and critical thinking
Reporters polled in this survey rated about half of PR content as “good,” however, many noted a “significant variance in targeting and editorial depth.” Indeed, original thought and critical thinking are crucial.
One coincidental and recent comment from a comms professional seems relevant here:
“Stop hiring comms people who talk and write like corporate robots,” said Christin Culver on X. “They will lead you astray. Hire comms people who are culturally attuned, slightly unhinged, and witty will lead you to the promised land.”
I’d add that clients and employers who want to be ‘positioned as thought leaders’ shouldn’t ask PR to write in isolation. Even seasoned PR professionals need direct access to customers and executives to make this truly happen.
As I’m prone to say, thought leadership actually requires thought and leadership.
I first heard about this study from a post on MarketingProfs: How journalists feel about the use of AI by PR professionals.
Subscribe by email for free:
Check out my weekly blog posts, weekly podcasts, or a monthly newsletter via Substack that rounds up interesting reading from the last 30 days (examples).
If you enjoyed this post, you might also like:
Is the influence of traditional media – and by extension media relations – waning?
Image credit: Google Gemni and respective study
