In multiple ways, Artificial Intelligence is redefining the role of the public relations professional. Some of that change is the result of new tools that automate processes that once consumed copious amounts of time. One such tool reviews services that solicit expert commentary at journalists’ requests, then crafts responses. The marketing of this tool, dubbed Synapse by its Lithuanian founders, has sparked a considerable amount of controversy over ethical considerations. Neville and Shel discuss the pros and cons in this long-form FIR episode for July 2025.

Communicators are now also supposed to be able to detect phishing attacks disguised as media inquiries, to abandon age-old metrics in favor of meaningful outcomes, and overcome old tropes, like one wheeled out by former communicator Melinda French Gates, who claimed without evidence that tech executives like Mark Zuckerberg have aligned themselves with the Donald Trump Administration only at the behest of their communication teams.

Also in this episode:

Links from this episode:

Links from Dan York’s Tech Report

The next monthly, long-form episode of FIR will drop on Monday, August 25.

We host a Communicators Zoom Chat most Thursdays at 1 p.m. ET. To obtain the credentials needed to participate, contact Shel or Neville directly, request them in our Facebook group, or email [email protected].

Special thanks to Jay Moonah for the opening and closing music.

You can find the stories from which Shel’s FIR content is selected at Shel’s Link Blog. Shel has started a metaverse-focused Flipboard magazine. You can catch up with both co-hosts on Neville’s blog and Shel’s blog.

Disclaimer: The opinions expressed in this podcast are Shel’s and Neville’s and do not reflect the views of their employers and/or clients.

Raw Transcript:

Shel Holtz (00:02)

@nevillehobson (00:15)

Shel Holtz (00:19)

Starting with a review of the episodes since our last monthly. Neville, you want to give us that rundown?

@nevillehobson (00:50)

Yeah. So starting with the last monthly, 469 on the 23rd of June, the story that featured in the show notes was, is internal communication failing? And that topic was on a growing body of research suggesting that employees are more disconnected than ever. It was a good discussion. Other topics?

of the of the six we covered included social media has overtaken television as Americans primary source of news. And Pope Leo the 14th has called for an ethical AI framework in a message to tech execs gathered at the Vatican. That’s a significant topic, which will become clearer in the next few days, I would say I’ll talk more about that in a second. But we’ve got a comment there too, don’t we, Sean?

Shel Holtz (01:38)

This is a crucial point for the communications profession. Comms people operate at the intersection of tech and user behaviors. There’s a danger here. We might start convincing ourselves that just because we can predict reactions with AI, it’s all right to use predictive analytics to manipulate the employees we are communicating to. And I am totally with you as to the need to frame AI, not as a product, but as a public conversation, particularly about the human dignity of the users. It is indeed exciting to have a pope.

so well versed in tech and math.

@nevillehobson (02:39)

Shel Holtz (03:02)

@nevillehobson (03:13)

Shel Holtz (03:16)

@nevillehobson (03:38)

Can you be influential and anonymous at the same time? We asked. There’s a new brand of influencer. We explained faceless creators who wield their influence while never appearing on camera. Look at the pros and cons as we did. And there’s a comment on that one too, Shel.

Shel Holtz (04:22)

@nevillehobson (04:23)

Shel Holtz (04:46)

It also allows for faster production in some cases too, if you don’t have to put your face on camera and worry about lighting, camera quality, audio, et cetera. And that’s pretty much what we said in the episode too, in terms of why people are doing this. They have a full-time job in a lot of cases, and this is a side hustle and they really don’t want their name in their face out there. So they come up with…

another approach to being an influencer, often through an animated figure, which has gotten a lot easier to do these days, thanks to AI.

@nevillehobson (05:47)

Shel Holtz (05:54)

@nevillehobson (05:57)

Yeah. All part of the change. So 472 on the 16th of July, that was about the evolution of trust. We talked about new research that reveals that B2B decision makers have increasingly recognized the importance of trust. And I must say, Cheryl, we did make the comment ourselves, are they still?

talking about this thing as if it’s a new thing. It does surprise me so many times when I see that. But we analyzed the data and explored some of the opportunities for communicators to enhance organizational trust. We have a comment, right?

Shel Holtz (06:33)

from transparent communication and authentic storytelling to consistent branding and content that delivers on its promise that can genuinely create customer confidence and loyalty.

@nevillehobson (07:13)

Shel Holtz (07:33)

A fascinating one and had a long discussion about it with a colleague at work. This goes back to this notion, I think, of message mission control, which I first heard from Pitney Bowes probably 22, 23 years ago. And it’s all about this notion that there are elements of the technology experience that really aren’t in IT’s wheelhouse.

@nevillehobson (07:52)

Shel Holtz (08:07)

I said, really, does your printer approve everything that goes out to all employees as a printed memo? So, yeah, it’s a fascinating topic and one I actually agree with. I’m not 100 % convinced that it should belong in internal comms. I think maybe an ad hoc committee, an ad hoc task force that includes IT and comms and maybe a couple of others.

maybe users, but absolutely, internal comms should have a seat at that table.

@nevillehobson (09:01)

headline, I suppose, that gathers it all together as AI and human dignity. But it was quite a conversation we had. So you, I, you and me, plus, as you mentioned, Sylvia Cambier as our guest co-host, Sylvia was instrumental in securing this interview for FIR. And over a period of 40, 40 or so minutes, we talked with Monseigneur Tai that was centered around a document that the Vatican published earlier this year.

It was called Antiqua et Nova, and it is a reflection, if you will, on AI, in contrast to comparing it to human intelligence. And it focuses very strongly on ethical responsibilities tied to technological development and the Church’s mission to guide AI towards the common good. That’s summarizing it very, very simply.

But lots of topics within the broad subject matter. What makes human intelligence distinct in the age of machines and algorithms was one of the points we touched on the ethical responsibilities of those designing and deploying AI. Where does that sit? The impact of AI and dignity justice on the workplace. We had quite a few questions about the workplace, by the way, that are very pertinent as this, you know,

this technology, if you like, gets deployed throughout more organizations. And also, I guess, a point of helping understand why the church is involved in this, unique role in the global conversation on technology and humanity. So it’s not all about the tech, what about the people? So that was a really, really good conversation. So that will be published on Tuesday in around about 830am.

UK time, so that’s GMT plus one. And it’s definitely worth a listen, so make sure you do not miss that.

Shel Holtz (11:15)

maybe three levels down, which we certainly have in larger organizations, people leading teams, but also reporting up through the communications function. So that featured Russell Grossman from the UK, Sue Heumann from Canada, Mike Klein from Iceland and Robin McCaslin from here in the U.S. Great conversation. It’s up on the FIR Podcast Network right now. The next episode of Circle of Fellows is going to take place at

4 p.m. Eastern Time on Thursday, August 21st. The topic is on sustaining sustainability. How do we keep it front and center in our organizations? The panelists for that one will be Zora Artis from Australia, Brent Carey from Canada, Bonnie Kaver from the U.S., and Martha Muzyczka from Canada.

So that’ll be a fascinating conversation. There are definitely headwinds on the whole sustainability front and the whole corporate social responsibility front. So looking forward to that one. And now that we have addressed all of our housekeeping issues, it’s time to jump into our reports right

@nevillehobson (12:50)

Synapse Media has said it has instructed its lawyers to take action against Wellstone. So you get a sense of where that’s all going to go. But here’s how Synapse works. It reads a journalist’s request for comment, scours online sources, including books, reports and podcasts, and then uses generative AI to create two plausible, personal sounding responses. These include quotes and even made up anecdotes, all written to sound like they came from a real human expert. The PR professional,

simply chooses the version they prefer and copies it into an email. Wellstone claims this approach can quadruple output. They say one person can now do the work of five. The tool has been marketed with a one-time purchase price of $2,500. And while it’s still early days with no known customers, the ethical alarm bells are already ringing. Critics have called it unprofessional and deceptive. Alastair McCapra, CEO of the CIPR,

when so far as to say Synapse represents the very definition of unprofessional conduct. And in a comment on the press gazette supporting about this, Harrow owner Brent Farmelo said, every query on Harrow goes through AI detection and we permanently ban users who rely solely on AI to generate responses. The concern is that Synapse undermines trust, both in PR and journalism, by introducing fictionalized AI generated content

into a process that relies on authenticity and credibility. What’s more, it risks creating a layer of misinformation. If AI generated quotes are published as if they came from real people without fact checking or disclosure, the implications for journalism, public discourse, and the integrity of media coverage are quite serious. So the big question is this, as generative AI becomes more capable and more accessible,

Should PR professionals lean into automation for efficiency or are we crossing a line where convenience erodes trust?

Shel Holtz (15:16)

@nevillehobson (15:17)

We did, a couple of years ago.

Shel Holtz (15:44)

what they have written and tailor a pitch. then same as with Synapse, it will allow you to copy and paste that pitch and send it to that journalist. This is automating a process. It improves your productivity and your output and nobody complained about it. I can see some of the concerns around Synapse, but on the other hand, we talked about Harrow, which

Peter Shankman started and then sold. He’s got another one that he has started because he didn’t like what happened to Harrow. But in any case, they’re banning people who purely use AI. Well, if you’re a PR person copying and pasting the output from Synapse and simply sending it, then being banned is on you. You should never do that with any output from AI, whether it’s a…

an email you’re having at Help You Write or an article you’re having at Help You Write, whatever you want to go through and at the very least heavily edit that to make sure it sounds like you, to make sure it says what you want it to, to add anything that’s missing, to delete anything that you don’t want going out. You would do this with an intern, right? You wouldn’t just send what an intern gave you. So I think that’s, I think bad PR practitioners.

@nevillehobson (17:21)

Shel Holtz (17:27)

And again, if you’re editing it to make sure that it does represent what you want to pitch, it should be fine. The one part of your report that made me sit up though is what they’re charging. Was that $2,500 a month?

@nevillehobson (18:05)

Shel Holtz (18:07)

@nevillehobson (18:28)

perhaps this is hence the, you know, hurry, hurry, this is a short term one time deal, get it now before he goes away type of approach. Interesting what you said. I thought about profit in passing, but didn’t really give it any. No, this is not the same thing I said to myself, but you’ve introduced it. So.

What I heard what you were saying was that this is the same thing as profit. I don’t think it is.

Shel Holtz (18:53)

it’s not the same thing as profit, but it’s automating a PR process and.

@nevillehobson (18:58)

without anything else, which is what this can do as well. This can also do, where you can edit it stuff if you want to, but it’s actually being pitched as you do not need to do that at all. And it’s being pitched as this little quadruple output, output. One person can now do the work of five. Ergo, you don’t need those other four people, right? Get rid of them, that’s save money. So I sense, I see the alarm centered around that. I’m not sure I agree with the criticisms about

basically saying this is unethical and we should not allow this to happen, et cetera, because, wait a minute, is profit like that? And I believe they are an ethical business. We interviewed the founder and have a high regard for him. I keep an eye on what developments are happening with profit. So I know a number of people who use that service. This is not the same thing it seems to me. Yet, your dismissal of it, if you like, is a valid point to make, I think, in this time, because

Things are moving so darn fast with AI that what you described with something like a chat GPT AI agent or whatever else is next around the corner is going to challenge this like nobody’s business. So who is the market for this? I wonder the company’s based in Lithuania. According to the press gazette, they’ve been on an email blitz around the agency world in London seeking sales, effectively the sales pitch emails.

The writer at PR Press Gazette got hold of one of those and the report, which is linked in the show notes, dissects that quite severely. So I can understand Alastair McCapper at the CIPR that saying this represents a very definition of unprofessional conduct. I wonder what Alastair would say about profit. I’m not challenging Alastair for anything at all, just wondering out loud in light of what you and I are just discussing.

Critics say this is unprofessional and deceptive. If the similarity with tools like profit are such that you give it the prompt, as we now call it, you prompt the tool and it gives you a response. You copy that and stick it in the email and send it out. So is that unprofessional and deceptive? To me, this is the heart of the wider debate about this, that embraces using generative AI.

write content and should you disclose it and stuff like that. I struggle in understanding those criticisms, frankly, because what’s the difference between a prompt to chat GPT and getting a response, then let’s call it a prompt to your contractor or your intern or your colleague. Hey, Jim, can you draft a first draft press release on this subject in this five bullet points? What’s the difference?

you’re using someone as a tool to assist you. And in the sense of this, you’re using this tool as a means to assist you in achieving your goal. That said, equally can see why it attracts such powerful emotive responses, because this is after all, all part of the wider picture about the robots are taking over, am I gonna lose my job and all those kinds of things. So I get that entirely.

And you can’t pacify that sort of worry by saying, this is a tool, that’s all. And it’s nothing to do with deceptive practice, because that’s not the perception that people might have. So I don’t know where this will go. Are we crossing a line where convenience erodes trust? And to many, the answer to that is yes. To others, maybe not. So I don’t know where this leaves its shell, to be honest.

Shel Holtz (22:58)

People can use tools deceptively and it doesn’t please me to know that they’re marketing it based on your ability to use it deceptively, but you don’t have to, right? You can abide by the code of ethics from PRSA or IBC or CIPR or what have you and use this ethically, but what is it doing really? I mean, I get Peter Shankman’s new version of Harrow. I subscribe to that.

So this would just read it for me. I do this with a service called Drip with all the AI focused newsletters that I subscribe to. I don’t have the time to read them all. So I have Drip get the subscription and I get a daily summary of everything that’s in them. That’s essentially what this is doing is going through these services and identifying journalists looking for quotes or interviews.

from something related to your client base or your organization. I’m looking for something where I can get our construction subject matter experts and thought leaders into the press. you know, occasionally something shows up. It just reads it for me. And then it goes out and it finds information that we can use to pitch that. Now I’m going to align that if I find it, if I were using something like Synapse.

I’m going to align that with our area of subject matter expertise and the work that we have done and the people that I would reach out to to be the interviewer, the source of the quote. All it’s doing is streamlining. And is it maybe going to lead to the need for fewer low level PR people to do all of this work? Maybe that’s where we’re headed with this. I think we have to deal with that and and associated with that, by the way.

we have to deal with how are we going to bring people into this profession if the entry level work is being shoved off to AI. But I absolutely see it. I I’ve used the OpenAI agent to identify podcasts that I should be pitching, recognizing the importance of that in AI optimization. I want to get our subject matter experts onto podcasts. So I said, go to our website.

understand our markets, understand our areas of specialization, then go find podcasts related to construction, to the AEC industry, and to the markets that we serve, and visit their websites for those podcasts, find out if they have a policy or a process for pitching guests, and then put it all in a spreadsheet for me. And I just sat back and watched while it did it.

@nevillehobson (25:55)

Shel Holtz (25:56)

know, that’s something that I would have given to a lower level employee at some point in my career. Now I type a prompt. So deal with it. This is coming.

@nevillehobson (26:04)

it is. I think, though, there’s a couple of clues in the press gazette’s report, which I which I use primarily to come up with my own thoughts on this, particularly the part where I said in response to what I what I found out if I generated quotes are published as if they came from real people without fact checking or disclosure, then there are serious implications. Well, yes, but that applies to

everything you see out there if people do that. They take something generated or worse, someone else wrote it and let’s pass it off as their own. What’s the difference between that and an AI doing it? So that’s down to people’s behavior. The concern that Synapse undermines trust, as Press Gazette’s report talks about, by introducing fictionalized AI-generated content into a process that relies on authenticity and credibility. I stopped myself when I read what

they were saying and what the Lithuanian PR agency was saying too, that in fictionalized AI generated content, I don’t think that’s what people would be doing fictionalized, meaning it’s not fact, it’s fiction. I’m not sure what that looks like.

Shel Holtz (27:13)

is the AI getting its information from? It’s from the training set.

@nevillehobson (27:17)

Yeah, exactly. there’s a lot of questions about this. Yeah, I do understand that the alarm bells that is ringing nevertheless. So how would how would they address that? I don’t know. think you PR professionals need to be very clear on what resources they’re using to do this. Follow those codes of conduct that you referenced. But this surely does come down to

belief in you’re doing the right thing, are you not? Or are you doing what people are alarmed about? Just copying and pasting and passing it off as genuine. That is in that case, you’re not a professional at all. So this is this is part of the shifts that are happening show that alarm the hell out of a lot of people. Yet this is the landscape, we have to figure it out.

Shel Holtz (28:06)

joined a professional association without having signed on to a code of ethics. And they do whatever they think they have to in order to produce results for a client, regardless of how unethical it is. And they’re the ones who are going to use this the way Synapse is pitching it. So this is a broader problem. The rest of us can use these tools, you know, sign on to the Venice Pledge, the…

the pledge that was developed by the Global Alliance around ethical use of AI by public relations practitioners and communicators. Pay attention to your own code of ethics and use these tools ethically. But I don’t see a problem with what Synapse is doing. I see a problem with the way they’re pitching it, but I don’t see a problem with the technology. I don’t see a problem with using this in an ethical way.

@nevillehobson (29:20)

Shel Holtz (29:25)

more critical and I’m going to talk about that a little bit more later in the show. But let’s start with phishing attacks from fake journalists. Yeah, you heard that right. Recent reporting from Axios confirms rising incidents where criminals impersonate reporters. They send what seem like legitimate interview requests or pitches to PR pros and clients alike. These scams jumped around 17 % since last September fueled by AI’s ability to assemble credible sounding emails.

using personal and outlet-specific details. These journalist phishing emails are not mass spam. They’re highly targeted, AI-generated messages that lure PR professionals or executives into revealing sensitive information or clicking links. In one case, attackers impersonated major outlets to pitch exclusive coverage, weaponizing publicly available data, like publication history and colleague names, to appear legitimate.

Now, the result of this is that PR teams are having to act as security filters as much as media liaisons. As experts advise, when a suspicious DM or email arrives claiming to be a journalist, the best defense is polite paranoia. Don’t click, don’t respond directly, forward it to your media relations team for vetting and verify the pitch through trusted context. Beyond phishing, PR professionals are also debating measurement frameworks.

and facing pushback on yet another measurement metric. Katie Payne, the industry’s measurement queen and a longtime friend, I got to know Katie really well through my involvement with the Society for New Communication organization. You and I were both fellows. right, Society for New Communication Research. Katie was heavily involved.

@nevillehobson (31:31)

Shel Holtz (31:36)

This critique resonates with ongoing efforts to move past outdated PR metrics. A group of major agencies recently announced they’re ditching impression metrics entirely in favor of readership-based analytics, thanks to platforms like Memo, which track actual article consumption and engagement across 10,000 publications. But this transformation requires education. Clients and executives are used to big numbers, even if those numbers don’t reflect meaningful impact.

Making the case for smarter outcome-driven communication means resetting expectations and elevating conversations beyond clicks or column inches. Anybody actually still measuring column inches? Finally, let’s touch on reputation fallout from misaligned values. Melinda French Gates, the ex-wife of Bill Gates and co-founder of the Bill and Melinda Gates Foundation.

⁓ recently called out tech leaders who publicly support public political figures. And what she did was she claimed that they’re following advice from their communication teams. Now, when someone with that kind of stature starts reinforcing old tropes about PR, it just amplifies the perception. And it was particularly shocking considering that she worked in communications before she met and married Bill Gates.

So the third headwind PR pros now face involves values misalignment and reputational risk in high stakes public discourse. Melinda Gates, she’s a billionaire philanthropist now, recently criticized tech CEOs who have publicly aligned with former president Trump. This would be Mark Zuckerberg, for example. Her point is that some leaders appear to be pivoting away from their values not out of conviction,

but because a communication advisor told him it was the right thing to do. Here’s what she said in a podcast interview, word for word, we have all time should be living these values out and not pivoting to what some comms person tells us. Now this statement lands as a rebuke and caution for PR professionals. Authenticity matters more than it ever has. If executives act in a way that contradict their stated beliefs because it’s convenient or media safe,

That gap is increasingly noticed. Crisis communications today can become long-tail reputation erosion, but it’s also a reinforcement of an old negative PR trope. And when someone of Melinda Gates’s stature says it, it amplifies it tremendously and it doesn’t do us any good. So just to recap, we got to look out for phishing attacks that are aimed at PR people. We have to look at the measurement.

that we’re using because increasingly the old metrics don’t work. And we have to watch for these claims of what PR is doing that is harming society, which is I just can’t believe that a PR person is out there counseling their CEOs to align themselves with any political figure in particular.

So taken together, these trends suggest that PR’s future is just about placements. It’s about credibility, ethics, and strategic integrity.

@nevillehobson (35:16)

certainly here in the UK from the the cyber security folks in the government, the government agencies who address this, particularly in light of awareness is high now. Following the cyber attack on Marks & Spencer, the big retailer that took them off that took the literature, the website down for ordering for three months, it’s cost of serious money. And it’s had a massive impact on their credibility to the extent that

latest league table show they’ve fallen from the top spot as the most respected retailer in the UK. For years, they’ve held that position. They’ve now dropped. So Waitrose or John Lewis, another big retailer, their supermarket Waitrose. So they usually be jockeying a position like this between the two and they’ve now taken the helm. And their troubles, by the way, are still not concluded. But this is interesting. Fishing in temperature on the rise,

Shel Holtz (36:14)

@nevillehobson (36:31)

just a quick bit of research made me even more suspicious. So I did nothing with it, but I’ve now learned that that was a phishing attempt. So you got to be on the ball with this. And I think tying it all up is this with the others you mentioned, and particularly the I did this because a commerce person told me to you think we in the profession need to get a grip on these issues. Is it do you think

Do we need just more training of people with that kind of thing? Certainly with phishing, this requires, I think, more than the profession is currently doing to alert people from two perspectives. One, the communicator as the counselor and advisor to the organization what to do about these things. But also as the more proactive role in helping probably more senior people, but probably everyone in the organization to

know what to have, what to do if you think you’ve been the victim here. And I’m not talking about these kind of slick, animated videos that a lot of companies tout and push out when when you they want you to do this, you know, online training and get awareness and learning about how to recognize these things. They I often wonder what true value do they have in helping people understand this issue of what to do.

So there’s that opportunity for communicators. But I do wonder though, with this, the other things you’ve talked about, that do we have the right skills in the profession, in the profession, to help communicators truly know what to do in these situations, or not? I wonder.

Shel Holtz (38:35)

require people to do annual online training. And I think the good that it does is at least it creates some awareness for a while, but it’s a requirement. We can monitor who has taken it and not and reach out to the people who haven’t to say, you you need to take this. And then if they still don’t, we can reach out to their managers and say, you have people who haven’t completed the training. We do this around safety and workplace violence and workplace harassment too.

But it’s not enough. You need to keep a drum beat up. First of all, as new techniques emerge that we are aware of, we communicate those very short little articles and we get them out through multiple channels. We have our executives talk about it at the all hands meeting. We bring it up occasionally at the monthly operations meetings. And with all of this, even digital signage, really quick hit awareness.

campaigns. Even with all of this, there are employees who are not going to hear it. One thing that we’re doing to try to make people pay more attention is to say, everything that we’re talking to you about that can affect us here at our company can affect you at home too. So the same techniques are being used to get credentials from your personal accounts. so pay attention to this because it could save your

personal life, not just our company. So yeah, you’re right. We have to become conversant in cybersecurity and stay up to date on the latest trends, the latest data, the latest techniques, and find ways to break through the noise and to capture the attention of our audiences to make sure that they know about this. I have not seen IABC PRSA, PR Week, or anybody else talk about this.

fishing yet. This was an Axios report where I found it. So we need to get the word out.

@nevillehobson (40:53)

Shel Holtz (41:10)

@nevillehobson (41:16)

but as content designed for AI systems. Her argument is straightforward. Press releases today serve two audiences, she says, traditional media and generative AI platforms like ChatGPT, Claude, Perplexity, and Google’s AI Overviews. That means a press release isn’t just a story, it’s training data. It needs to inform and persuade not only human journalists, but also algorithms that summarize, reinterpret, and surface that content in response to user queries.

And for that to work effectively, releases need to follow a structure that’s both human readable and machine digestible. Evans suggests that the modern press release should still follow journalistic standards. For instance, AP style, clear sourcing and strong headlines, but also include structured formatting, clear language and trust signals that help AI tools understand and elevate the content.

On the points of style, by the way, if you’re in the UK, you’ll likely be using the Oxford Guide to Style, formerly known as Hart’s Rules rather than the AP Style Guide. Doesn’t matter actually, as long as you’re using one of those, that’s good. That might mean including prompt-like phrasing, definitive outcomes, and text that aligns with how AI systems pass authority and relevance. This approach recalls an earlier experiment. I mentioned this is coming at the beginning here.

The social media press release, the template introduced by Todd Deferin back in 2006, it reimagined the press release for the Web 2.0 era. Multimedia, bullet points and links instead of dense paragraphs. That idea didn’t gain as much traction as many hoped it would, perhaps because the infrastructure wasn’t there yet. But in 2025, the infrastructure most definitely is, says Evans. AI tools are reshaping.

how people find, consume and trust information. Search engines are incorporating AI generators, summaries, news aggregators and research platforms use LLMs, large language models. If your press release isn’t visible to AI or isn’t written in a way that helps its surface, it might as well not exist, she says. So the conversation now is this, should we be writing for the journalists and the algorithm? And how do we do that while still maintaining journalistic integrity

clarity and credibility. There are challenges. Here are three. Lack of standards. There’s no defined schema for AI ready press releases yet. No equivalent of schema.org for PR. AI system opacity. We don’t fully know how LLMs weigh or interpret content from wire releases. And skepticism. As with social media press releases, some PR pros may dismiss this as overkill or gimmicky without clearer benefits or just

don’t care about it at all. So is it time communicators embraced this dual purpose writing style, or is it just another passing trend? Shell, what do you think?

Shel Holtz (44:38)

We’re going to talk about this in much greater detail in the next report, but just to telegraph that a little bit, what AI training sets look at and find credible is third-party reports, not what you publish on your own website. Most companies do have a repository of their press releases, we do, and it goes out on the wire from you, but what the AI looks at for

the sake of credibility is what publications have picked up that press release and written about it. And of course there are publications that will just publish your press release the way it is. They’re just looking to fill space and have stuff on their outlet. But the good ones are going to use it as a basis for original reporting. So it needs to be a good enough press release to entice that type of reporting. ⁓ For those who

just publish the press release the way it is, it’ll be really important for that press release to take a different approach to the point of your report. ⁓ One of the things that I understand AI looks for, the LLMs when they’re extracting the press releases and the content that the press releases were used to create from the media outlets,

What they’re looking for is less of the what we do and what we sell and what our products are and what our services are and more of the how, more process. So I think that’s a shift that we have to make in the angle that we’re taking when we’re crafting press releases. And I think that AI and this whole direction that we’re going,

has reinvigorated the press release. mean, people have over the last few years been talking about it as a relic and why are you doing this? You need to be using influencers and you need to be doing content marketing and press releases are so 50 years ago. AI has reinvigorated the value of the press release for just this reason.

@nevillehobson (47:06)

complimentary thing called the algorithm that we need to create content for. And Sarah Evans actually included in her LinkedIn post, a rather neat LinkedIn style slide deck, you can’t download it or grab it, you can only scroll through it. That showed the schema structures that she proposes that could be where we go next with this to make it usable by AI systems. And I think that’s

That’s the bit that is key to this, is key to understanding the possibilities of this. So, and I like what you said earlier too, that the profession’s responsibility is to develop the schema for this. And that could be a great opportunity for professional associations to take a lead on something like that or one in particular. So that you have something that’s universal. And once you’ve got that, which I guess supports my feeling that

social media press release one of the reasons and may not be the most significant reason why it didn’t take off is there was no infrastructure to support it or making it easy for people to adopt it. I remember seeing back then version 2.0 of the template it didn’t get beyond 2.0 but when I saw 1.5 I then saw variations of that all over the place as people were adapting it to the knees then sharing what they did so utter confusion everywhere as to which was the valid one to use.

That I don’t think is likely to happen with this. The landscape’s completely different apart from anything else, but maybe the timing’s right.

Shel Holtz (49:00)

You don’t need infrastructure for this. All you need is a different approach to what you’re writing. You don’t need to create sections and have links to things that are uploaded. I remember Jason Keith, think his name is, had a press release distribution service that was focused on the social media press release. I don’t think it’s still around, but this is just the way you write it. It’s not a different…

@nevillehobson (49:07)

Shel Holtz (49:28)

@nevillehobson (49:43)

many, many others, New York Times, although I got a subscription to that, but I could do this if I wanted to. The only thing the only one it doesn’t work with that I discovered is PR week, I have no idea why this cannot get into that one. they are pretty good at it, at stopping people bypassing the paywall. But it’s a very interesting one. And it works all the time. It works flawlessly. Most of the other time. Also to mention you when you talked about

the Online Safety Act in the UK and verifying age to protect children from pornographic content. The UK communication regulator published a pretty neat explainer on how age checking works in the UK. It’s worth a look at. I’ve noticed a lot of comment in the last few days about people saying how easy it is to bypass this verification check. And so if that’s the case, they got to step up and fix that. Blue Sky,

introduced this this past week in the UK as part of this law. So the other day when I logged in in the morning, I was got a pop-up say I had to verify my age. My first thought was, well, you know a lot about me and I’m verified with you, so you know my age, but I get it. They have to go through this process. The third party provider that does it. You’ve got to photograph your driver’s license or a passport or some other government ID.

upload that file. And then the next step in the verification is either a really call it a picture of you that you share or check on a credit card. I chose the latter because that was the easiest one to do. And so I’m verified in that sense. I just wonder though, how, how truly robust that is in light of the criticisms. But it’s good that you had the topic in your report.

Shel Holtz (51:56)

blue sky. It’s just for Pornhub and other porn sites. But it’ll be interesting to see how easy it is to get around those checks. So, I mean, you know.

@nevillehobson (52:35)

an interesting point to make because in the UK, the UK online safety isn’t just about porn sites. It’s related to any content that’s age restricted according to the new law that’s been passed. So there is that difference in that case.

Shel Holtz (52:49)

Occasionally, something will show up in a feed and I want to read that article maybe because I want to write about it. Maybe I want to include it as a topic on FIR. And I can’t get to it because of a paywall. I like the sites that give you five or 10 free articles a month before the paywall kicks in, because that would be enough for me for the Financial Times, where I get maybe two or three articles a month that cross my feeds that I want to go.

take a look at so that I might incorporate that information into a report or even alert my leadership at the company I work for to this information. So I think this rush to paywall everything is, I understand it, but I don’t think it’s well considered the approach that’s been taken so far.

@nevillehobson (53:57)

Shel Holtz (53:58)

Don’t just link to your content anymore. They generate answers that borrow authority from trusted sources. And who’s always built trust at scale online? That would be us, PR and communications professionals. Research from Search Engine Land underscores that AI-powered search engines increasingly rely on brand mentions, reputation, and authority signals retrieved from across the web.

That means that when AI answers replace blue links, it’s not keyword density or meta tags doing the heavy lifting. It’s earned media mentions and narrative consistency. Ginny Jitrich wrote about this, and she puts it bluntly, PR has always been about building trust and carrying that trust into every story or quote your brand participates in. Today, when AI synthesizes answers, it looks for those signals. And as she writes,

We’re telling the same clear story everywhere, so AI can’t miss it. That’s why disciplines like GEO, that’s Generative Engine Optimization, and AEO, that’s Answer Engine Optimization, are emerging. GEO aims to ensure your content actually gets cited to become part of the AI’s answer. AEO focuses on structuring content in a conversational, Q &A-friendly way so that AI can…

easily use it. And this is probably what’s going to find its way into a schema if a professional association gets around to sharing that. But no matter how well designed your website is, AI models still prioritize third party editorial sources. And of course, as I mentioned before, a lot of people who get a press release are going to use it as a foundation for their own original reporting. Muckrax 2025 report.

highlights that over 95 % of AI-cited links are from earned media, not owned or paid content. And these citations actually reshape the entire answer AI gives. If your PR team isn’t securing coverage across high-quality publications, journalistic outlets, podcasts, analyst quotes, your brand won’t surface in those AI-generated answers, no matter how much of it you’ve published on your own sites. For example, when people ask generative AI something like,

best web design firm in Chicago, what really makes Orbit Media or others show up is repeated brand coherent mentions in reputable publications written by journalists consistent in message, quoted accurately. That’s what we do. And it’s the accelerator for AI visibility. So I’m increasingly seeing PR agencies evolving from just earned coverage into source engineering, designing narratives and placements that feed the machine.

As a recent piece puts it, strategic PR teams are no longer just telling stories, they’re engineering the data that AI models pull from. In practical terms, here’s what working PR teams are doing differently in 2025. First, they’re prioritizing earned media in authoritative outlets, not just for backlinks, but because AI is trained more heavily on sources like AP, Reuters, respected trade publications, industry podcasts. That placement leads directly into AI visibility.

Second, they’re applying semantic consistency across channels. The brand name, executive titles, and narrative themes are repeated. Third, they’re auditing the machine’s brand view, monitoring not just media coverage, but what AI knows about your brand and its outputs and adjusting messaging accordingly. Fourth, they’re structuring content for both people and algorithms. Clear metadata, Q &A formats, concise summaries that serve both readers and AI ingest pipelines,

And fifth, they’re geographically targeting their public relations for local news placements that can lift visibility and AI answers for place-based queries. For local businesses, this is a powerful, though under-leveraged lever. Meanwhile, TechRadar and others report that younger audiences rely on AI chat or conversational interfaces instead of Google, undercutting traditional SEO and forcing a rethink of discoverability strategies.

Brands now need structured, accurate, and syndicated data, often via knowledge graphs, to feed AI systems and prevent misinformation or omission. Startups like Profound, Athena, and Scrunch AI, got to love some of these names, have emerged to monitor and influence brand visibility inside large language model outputs, an entirely new tech niche built around this shift. And yes, PR agencies still matter, even with AI tools that can draft press releases,

because AI can’t build journalist relationships, craft narratives for trusted outlets, or manage timing and nuance the way humans do. Again, you gotta have the human in the loop. It’s just that AI lets you focus on these things. As one article put it, while AI can help with execution, it lacks the insight and relational intelligence that PR professionals provide. In short, public relations is no longer just a support for marketing. It’s the foundation of visibility in AI search.

@nevillehobson (59:33)

So she mentions that how AI is rewriting how attention, reputation, discoverability work, but the fundamentals haven’t changed at all. Trust matters, relevance matters, authority matters. These are the currencies of the AI era, she says, and they are the currencies of PR. She also mentions, and I was glad to see this, the PESO model in the context of all of this. I’ve seen quite a bit of criticism in recent months about PESO and how it hasn’t…

adapted to the changes we’re seeing in all this. I don’t agree with that, but I’ve not looked into it in depth, the comments that I’ve seen. But Jeannie notes that the Pace and Model campaigns are designed for human audiences and for LLMs to find, interpret and share accurately. So you don’t have to imagine for long, it’s already happening, she says, if you were wondering what is actually going on. AI is the interface through which people get information.

PR becomes the operating system that powers accurate brand aligned answers. Great visualization of that. can see that actually. Those who understand how to feed the system through trust, relevance, and strategic visibility will lead. So I think that’s a good take on the changes that we’re seeing happening in front of our eyes almost daily, that what actually matters. And it’s not about pivoting either. We just need to step up.

Shel Holtz (1:01:20)

They’re not reading what their professional associations are sharing. Somehow we have to get this information to people or, you know, frankly, what’s probably more realistic is that the people who do pay attention will succeed and those who don’t will fail and just fall off the tree.

@nevillehobson (1:02:06)

is about a shift in what social media is fundamentally for. Take Metta’s recent positioning. Mark Zuckerberg now describes Facebook and Instagram not as social networks, but as platforms for discovery, expression and entertainment. Kind of trips off the tongue, it? DEE, discovery and expression and entertainment. The implication is clear. Yeah, DEE, yeah. The implication is clear, reported French.

Shel Holtz (1:02:50)

@nevillehobson (1:02:55)

and algorithmically optimised to keep people scrolling. This represents a huge cultural pivot they say. Social platforms are now more like streaming services than networks. People use them not to connect, but to consume. And from a communication perspective, that changes everything. This shift is reflected in emerging trends identified by the drum and others. Brands are posting less, but focusing more on quality, short form video, strong storytelling and emotional tone.

There’s growing emphasis on social search. Users now search TikTok, Reddit or chat GPT before turning to Google, if they even do. Private communities are gaining ground. Places like Discord, Telegram and closed groups like LinkedIn offer engagement where the feed no longer delivers. And authenticity, or at least a performative version of it, is winning. Think of Duolingo’s chaotic TikTok presence when they killed off their owl logo.

didn’t phase anyone and their ⁓ presence on TikTok is in the millions following that, that didn’t decline at all. Think about the irreverent tone of Ryanair’s social media team. We’ve talked about Ryanair on this podcast before, Shell, and they insult customers, they do all this stuff and people love it. So their presence across social networks is increasing literally day by day. So that is a shift where you

normally would expect you to be totally aware of treating with the customer. The customer is always right. And those kind of differential relationships, not anymore. And people seem to like this. And there are two brands who are gaining from that. So this is not simply evolution. It’s reinvention. The feed is no longer a channel for personal expression or connection. It’s a content platform. For communicators, that means rethinking everything from platform strategy and influence selection.

to brand voice and metrics of success. So the question is this, if social media is no longer about being social, what is the role of communication professionals in shaping visibility, trust and conversation in the new landscape? Or put it another way, if social media is now more like Netflix than a network, what does that mean for how brands and communicators show up?

Shel Holtz (1:05:41)

was that social media was going to be a wonderful thing. The public commons giving everybody a voice. Boy, were we wrong. Were we naive about that? We did not anticipate the algorithm and getting people to doom scroll for every spare minute they have and even some minutes that aren’t spare. You see people at their desks at work doing that. So.

@nevillehobson (1:06:13)

Shel Holtz (1:06:26)

And yeah, when I do go to my feeds, interjected amidst those people that I do see whose comments to me or a post about something I’m genuinely interested in in their lives because of that personal connection, intermingled with all of that is what you said somebody characterized as the equivalent of streaming media. And how many videos do I end up watching because of that?

It is something where we need to make that adjustment and it requires being on top of the trends. And this trend isn’t going to last forever, forever either. I mean, nothing does. So especially with the shifts we’re seeing in media. So it is incumbent upon us to stay on top of these changes, even though, you know, I mean, this has been this has been occurring over time. This isn’t something that was a sudden

veering of the approach that the social media outlets have taken. This has been a very gradual shift that we now find ourselves at. I don’t want to say the end of it because it is probably an ongoing change, but we find ourselves where we are. And if we want to produce the kind of outcomes that our clients and our employers expect of us, we have to accommodate them. Otherwise, we’re just talking to ourselves out there.

@nevillehobson (1:08:08)

not everyone else does like that, but they’re benefiting from it in their niche with their type of audience. mean, there multiple questions arise from that. it ethically right thing to do? What about others in the organization? How do the employees feel about it and things like that? So I think it would be difficult to say, to answer the questions, what does this mean for our brands and communicators show up, show up, get to understand how it works and

what you’re going to do about it, what it means to you and your brand. I mean, I can’t think of a simpler way of saying that. And certainly not going to sit here and try and say, here’s a detailed assessment of what you should be doing. You need to understand the shifts that are happening. You need to realize that this is transitional and this is not permanent in the detail. But the fact of this is likely to be long term. The old days that you described

with fondness of the past where everyone did engage and we welcomed it. We evangelized it 20 years ago. We wanted everyone to be online and sharing their lives. And like you said, Cheryl, I sometimes slap my forehead and think, what have we done? So it’s actually influenced me hugely in where I’m no longer online. And where I am online, what am I doing there that’s

very, very different to what I was doing 10 years ago, even five for that matter. So things are shifting. I’m an observer more than an active protagonist on this stage, simply because my interests and needs have changed from what they were five years ago. But I look at developments like this, I find this one quite fascinating as another step in the evolutionary movement we’re at. I think

The only thing I could say is precisely that. Understand how this is. Figure out how you’re going to behave in this and get to it.

Shel Holtz (1:10:37)

that fits the ethos of each of those networks where we’re participating. So it’s incumbent upon communicators to understand the content marketing component of their job. Part of that is perhaps moving away from social media or adding non-social media channels to the mix. So…

Picture the CMO, the chief marketing officer of a major fashion brand flipping through Instagram and thinking, is this really how I’m supposed to build a loyal relationship with Gen Z? So instead, she launches a newsletter on Substack and suddenly she has a sent to inbox channel where real conversation is possible. Brands are increasingly turning to Substack, not just email newsletters, but Substack in particular.

Not because it’s just another email tool, because it offers control, community, and credibility. According to Marketing Brew, in mid-2025, brands such as Madewell, Tory Burch, American Eagle, Rare Beauty, and The Real Deal launched newsletters on Substack. These aren’t boring corporate updates. They’re cultural briefs, contributor channels, founder stories, and stylist features that appeal to audiences in ways that social media can’t replicate these days. For example,

American Eagles Off the Cuff is designed like a group chat with Gen Z, culturally tuned in, conversation driven, and not overly transactional. Within six weeks, it hit over 2,000 subscribers, even though most of the content was not product focused. Now, you may wonder why Substack? Morning Consult found that Substack users tend to be heavy news readers, younger, they’re Gen Z and millennials by and large.

They’re highly engaged and more likely to trust brand communication delivered via newsletter than via social media threads. Substack delivers a clean, distraction-free experience and a built-in discovery feed encouraging deeper engagement. So let’s look at Rare Beauty. This is Selena Gomez’s brand, and it created Rare Beauty Secrets on Substack, offering behind-the-scenes stories from product development to mental health summits. It’s less about selling lipstick

more about sharing values. The newsletter is clearly resonating with high open rates and a Discovery Feed subscriber rate growth around 17%. Brands like Hinge are also experimenting with immersive narrative series like No Ordinary Love, a five-part literary series that tells moving stories tied to their dating app mission. The key is compelling content, again, content marketing that feels human and original rather than promotional.

So what are PR and communication pros learning from this wave? Well, the Content Marketing Institute lays out best practices that fit neatly with public relations instincts, establish a consistent cadence, use authentic storytelling, design for readability, integrate content contributors, and above all, make the content about value, not selling, first. Additional analysis underscores a few crucial points.

You can’t treat Substack like a traditional newsletter tool. It’s a community platform intended for editorial quality, not emails disguised as ads. Partnerships matter. American Eagle brought on Casey Lewis, author of After School, as a guest editor that cross-pollinated subscribers and raised credibility for their off-the-cuff newsletter. And founders have to be voices in these newsletters. Brands with strong founder voices. Schoolhouse, Gia and others.

They’re using Substack to create intimate access and narrative continuity. It works particularly well when a founder or a key personality leads the newsletter, though always anchored under the brand’s umbrella. There’s even an editorial economics side to this. While brands generally keep their Substack’s newsletters free for now as part of building trust, platforms like Substack are pushing tools to help creators monetize, and brands might explore premium options down the road.

For independent writers, ad revenue is already a growing part of platform economics, but most brands are focused on nurturing audience relationships first. Substack itself continues to evolve. As of early 2025, it surpassed 5 million paid subscriptions, and it’s doubled down on audio, video, and community features like chat threads, making it more than a newsletter tool, but a branded content environment. From a PR lens,

You got to consider that newsletters reinforce brand voices and values. They let you lean into storytelling purpose and humanity. They give you control. There’s no algorithm deciding how far they’re going to boost it up. No reach penalties. And your subscriber list is yours. It’s discovery driven. Readers come via Substax ecosystem, not just your current audience. It complements earned media. You can fund earned coverage insights into newsletter narratives.

deepen stories and convert that attention into subscriber trust, and it builds owned engagement. Comments, replies, feedback loops that signal two-way connection. It’s not just broadcast. Now, of course, there are alternatives to Substack. is one, and there are reasons to consider them, but Substack does have first mover advantage in a suite of tool that seems to be resonating with important stakeholder audiences. For PR professionals, brand newsletters on Substack

represent a return to relationship based communication that these changes in social media that we were just talking about have led to. So if you haven’t reconsidered your newsletter or whether Substack fits your audiences, you should. It’s not just for essays or creators anymore. It’s increasingly home base for brands that want to be heard, not just seen.

@nevillehobson (1:16:53)

WordPress to motor but that was a different platform. didn’t didn’t kind of accentuate that. I, I find from what I’ve researched myself even that the the you mentioned. Substack has first mover advantage. Yes, they did. But I’ve noticed that there’s a differentiator has emerged between places like Substack beehive is another one, and ghost of the type of person and what they talk about.

is attuned to one or other of the platform. Substack went through a phase early this year or late last year, I think, when they were accused of being anti-Semitic and they were hosting neo-Nazis and all this kind of stuff. People left them in droves. Many went to ghost. They probably recovered from that. But maybe that’s got a stigma that still surrounds them a bit, perhaps. find the thing that I find interesting too, Shell, about this in the bigger picture sense,

is this to me is like an evolution of blogging, of personal blogging that we had back in the day when CEOs didn’t do this really. And when one did, that was a big headline, when Richard Edelman stepped up. So the CEO of the Edelman PR firm and many others, this fulfilled something similar.

Shel Holtz (1:18:28)

Bill Marriott, remember, had a great blog.

@nevillehobson (1:18:32)

Merritt, absolutely. And of course, way before that, Bill, Bob Lutz, wasn’t it? Bill Lutz, Bob Lutz, General Motors, who was the kind of star writer in the very early days of blogging. The small block engineer used to write about a lot. I remember that. So, but I think the interest in this is the community building, you mentioned. This is not just something you send out, it’s designed to foster community.

And just speaking in my case, as a very tiny example as an individual blogger, I’ve noticed that with the newsletter that has fostered more engagement with readers. for me, what’s interesting is that’s happening more in direct personal feedback from a reader rather than the reader leaving a comment on the site via the newsletter, which is what you’d expect them to do.

I just haven’t figured out why that is yet, but I’m not too concerned because I offer a free newsletter is not a not a monetizable one. I’m not interested in doing that. And I’m gives me insights into the content. The point more than anything, though, is that this is a platform that is attracting more and more people like sub stackers, who traditionally would have potentially not shared stuff or if they did, it was via a blog or on LinkedIn or some other

closed group. Suddenly they are now controlling this on their own domain and building community on their own domain, which I think is an important element in all of this. Even some, I noticed some of the ones that are the handful I pay for. I look at some of the techniques they’re using to monetize, where they’ll offer something that is for paid subscribers only, and they’ll give you like 20 % of it, and you need to sign up to see the rest. Personally, I don’t like gimmicks like that at all.

But nevertheless, I get this is a different landscape than it was back then. So it’s good to see this. I have to say the growth of newsletters that are to do with building community, not just like you mentioned, just sending out an email as a marketing text. These are personal, engaging stories told by people who have a story to tell and do it in a way that people like and they’re willing to sign up for and in some cases pay for it. So I say.

more of this please on any of those platforms because this is what engagement really is about in my book.

Shel Holtz (1:20:52)

@nevillehobson (1:21:04)

Shel Holtz (1:21:13)

@nevillehobson (1:21:26)

Ha

Shel Holtz (1:21:41)

@nevillehobson (1:21:51)

Shel Holtz (1:22:00)

@nevillehobson (1:22:04)

Shel Holtz (1:22:29)

at gmail.com and we would love to play an audio comment if anybody would ever send us one again. We used to get a lot of those.

@nevillehobson (1:23:05)

Shel Holtz (1:23:08)

she’s interested in what we talk about anymore. Mostly Star Trek these days with Chris. Our next long form episode for August will drop on Monday, August 25th. So keep an eye out for that. But we will have short midweek episodes between now and then.

@nevillehobson (1:23:15)

Shel Holtz (1:23:29)

The post FIR #474: AI is Redefining Public Relations appeared first on FIR Podcast Network.