Name: Leveraging AI to Maximize Media Relations
Uploaded: 2026-02-26T19:16:09.471Z
Duration: 59 min 56 s
Description: Leveraging AI to Maximize Media Relations

Transcript for "Leveraging AI to Maximize Media Relations": Hi. Hi, everyone. Hey. What's up, Shelby? Look at all these people from all over coming in here. Denver, London. Holy cow. London. It's almost past past that person's bedtime. They wanna come to the live show. We love it. Hey, London. Awesome. Hey. Yeah. I see people that are, like, stuck in the snow. This is a perfect way to set yet another snow day. Right. Exactly. Well, not you. You're down in the sunshine. I I I will say that the high is 87 today where I'm at. So if anyone wants to come visit me here in Phoenix next winter, I you know how to find me. we'll do we'll do this session live in in person next time. Perfect. From the pool. Yes. Exactly. Awesome. Well, I know we have a jam packed schedule here today, and I also wanna leave plenty of time for the q and a. So we're gonna go ahead and get started. Super excited to go through today's session. As just kind of housekeeping reminders for everybody or for anybody who hasn't been on these calls before, we do have several user types, as we as our kind of comms professionals in here. So we have comms professionals who are representing clients, agencies, brands, in house, whatever that might look like. We also have individuals who are in here representing themselves and their own brands, kind of learning how to navigate that comm space by themselves without working with comms professionals. We do have some nonusers of Quoted. We also have users of Quoted. So, this session, we've set it up in a way where Gab here, who's on the call, she is going to talk us through utilizing AI for your overall comm strategy. And then at the tail end here, I'm gonna come in, and I'm just going to speak a little bit about how we can tie that into just your quoted experience overall. So you will get all of that information in there. If you were on last month's session, you probably saw how we really got stuck into the topic of AI on the call. So we have set this up very tactfully this time where we wanna make sure that we have enough time to go through the q and a, and we wanted to leave this a little bit more open forum for everybody to get a lot of questions answered. So we've left plenty of time towards the end for q and a. But throughout the call, as you have questions, you'll notice there's a q and a chat box. Feel free to pop those questions in there throughout the session. You will also see that there is a little upvote feature, and we highly recommend that you use this. So if there are questions that we can't get through, what we're going to do is we're gonna go through the ones that are being voted on most. So at that way, we can rank those, see which ones are the most pressing questions. So please upvote those throughout the experience. So I know I also promised last month that you guys were not just gonna have to listen to me talk the entire time, which is great news for me. I'm sick of hearing my own voice. So I am so excited today to introduce Gab Farrie. Gab is the founder of Off The Record, which is a community and coaching platform built for comms pros who are smart, ambitious, and just like all of you, trying to keep up with how fast everything is continuing to change in the comms world. So before launching off the record, Gab spent more than fifteen years leading comms at really high growth companies like Edelman, Salesforce, Slack, One Trust, Bumble, so on and so forth. So she has seen every flavor of crisis, executive requests, quick turnaround. She knows a little bit of it all. So we are so excited to have her here today because she's really going to help you understand what it's going to look like optimizing these workflows and figuring out how AI can make your job easier, but not just noisier. And she's going to be a really good resource because she's she's lived it all. So when we were chatting with Gab about this session, she said that she started off the record originally because she wished that when she was early on in her career, she had a resource like off the record. So that's why we've brought her in. We think that she's going to be a really good resource for all of you. So very, very excited for this. As I mentioned, we'll come in with the quoted aspect of this towards the tail end, and then we're gonna have plenty of time for q and a. And without further ado, Gab, I'm gonna let you take it away. Okay? Oh my gosh. Thank you so much, Shelby. That's. such a great intro. And when we were also talking before, you let me know that those who joined last month's webinar had so many AI questions that you could have done an entire session on AI. And my immediate thought was, like, challenge accepted. So thank you again for the intro, Shelby. Again, my name is Gobfari. For the last fifteen years, I've had this awesome opportunity to have a front row seat to what organizations look like when comms is at the leadership table and when it's not. And when I left Bumble, I had a choice. And the choice was I could go work for another name brand company, or maybe I could take all of the lessons from these last fifteen years of working with CEOs and boards and comms teams on driving business value. All that magic of, like, the one on one meetings with my teams and deliver at scale. And that's really what I'm gonna do here with all of you quoted and I do every day with Inside Off the Record. I believe that a rising tide lifts all ships. And the more that we can, as comms professionals, be better business partners for our executives, the more that we can be better as an industry. And so my goal is to help you earn your seat at the leadership table and work in a way that gets you home for the dinner table. Someone mentioned my sweatshirt. I do like to think of myself as your comms BFF. I'm in your corner to help you see around the corner. And a huge part of that today in 2026 is about AI. Right? So before we get into your questions, which I am so excited to tackle, I wanted to lay some groundwork on where comms is with AI. Now let me share my screen. Great. So This session is designed to be practical, tactical, and immediately applicable for you. And by the end of today, you're gonna understand where you stand on the AI maturity scale. You'll have some tools and tricks. But I'm also at the end of this session, I will send you an email with a sixty minute version of what I am going to be flying through today. So if anything felt too quick or you didn't get the whole thing, more detailed slides as well as key takeaways will hit your inbox hopefully later today. So here's how we got here. Last year, Axios HQ and I surveyed 171 comms leaders. Most everyone we talked to has ten or more years of experience. They were mostly directors and VPs and they split. Half were in house and half were agency or consultant. So this was an advanced crowd and probably looks a lot like a lot of people here on the call. These aren't just lurkers though in AI. Right? These were leaders. These are the types of people who are joining communities like off the record, responding to surveys, and on trainings just like you are today. And this is good because when I put together this session, I wanted to create a training that reflects where the leaders are in AI. And I hope that you'll realize you're not behind, but you're right on time to do more with your comms AI program. So the big picture is that AI adoption is universal. Maturity, however, is not. Here's the headline. 100% of the comms pros we surveyed use AI for work, but only 14% say that they are advanced users. That gap between maturity and adoption, this is your opportunity. Really quick, this is how we defined it, and I'd love for you to drop in the chat where you think you are today. Beginners are only about 8%. They're using AI for search or quick questions. The vast majority, 78%, are what I would call intermediate. They're using AI to write, edit, and brainstorm, but only 14% are advanced. These are the types of folks who are building workflows, custom GBTs, and automation. So let us know where you think you are today. But the fun part is that you can move up. And I'll talk in more detail about that in the training that I'll send you later today, but let's go on into our second insight. Maturity is the foundation, but it's certainly not the whole picture. Because once you understand where you sit in the AI maturity spectrum, the next question is, how are you using AI? And this is actually where the biggest myth got busted for me. People aren't just using AI for cranking out first drafts. The smartest comms leaders are using it for strategy. I thought comms pros would be leaning in AI for the rote stuff, like writing drafts, editing content, repurposing pros, those sort of easy wins. But that's not what's happening. Most of you have actually leapfrogged right over that and are using AI for strategy and higher order thinking, like brainstorming, messaging, pressure testing narratives. This is great, but it also opens up a huge opportunity because when you have AI help you focus on basic tasks, that is where you gain real efficiency. And until I did this survey, I actually thought everyone was using it for these rote tasks. Then my partner happened to come up to me one weekend and he said something like, ChatGPT is the absolute worst at writing. I was like, oh, what? Like, that's so embarrassing because I've been out here saying like, coms pros have to be better than just at writing because AI can now write for you. And so I dug into it with him and I realized that we were using AI in completely different ways. So I had fully trained projects on my tone, on my voice with tons and tons of previous samples of what I would call my, like, marquee communication samples. My partner didn't. I would always write my first draft. My partner didn't. I would always give my AI feedback. My partner didn't. I would always upload my final drafts into the project so it could improve. My partner didn't. So what I realized here is that we're using AI in different ways. So the way to really up level this growth opportunity of automating these rote content tasks, the more time you are gonna unlock for even more efficiency for your higher order thinking. So things like editing and content cleanup, these are where AI really can save you hours, but only if you've trained the system right and prompted it well. So I encourage you all to invest a little time in tone training and smarter prompting, and suddenly that rote work gets way faster, sharper, and more on brand. And to do that, let me tell you about the only prompt you'll ever need. You can literally just ask the tool, how should I prompt you for that? This alone will make you better. Now if this is too basic and you wanna go a next level, here's what I would encourage you to do. Take a project that you do over and over. Start here. How do I best prompt you to write the world's best media pitch for a project that I have coming up? It's gonna tell you I need to know everything about, you know, the journalists you're trying to track, the audience you're trying to reach, the product that you're trying to, you know, distribute to the world, etcetera. Where's, like, the news release, etcetera. You'll you'll talk back and forth. Go back and forth for however long you want. You finally get to your most beautiful end result. Now you're gonna say, write me a prompt that gets us to this end point much faster. Make it extremely detailed and have it ask me follow-up questions. It's gonna write you up something, then you're gonna take that. You're gonna put it into a project. Now you have built yourself a repeatable system. Now you don't have to start with the only prompt you ever need. Now you're gonna start with a project. So the next time you're doing media outreach, you can go right into that and already know so much about the way you work, the way you like to pitch, the way you like to write, the way you like to customize, the way you like tell product stories or whatever it is you're not seeing, etcetera. Okay. I go into all of this a bit more. I will tell you guys about it later. If this feels like too much, I promise you I will share more. So we've talked about maturity and strategy, and now I wanna talk about influence because you all know this, media relations is changing fast. I want to be very clear. Media relations has and always will be about reaching human beings. But now it's also about reaching machines. Machine relations is where comms pros are really stepping up as leaders in an age of AI because media relations is also becoming media excuse me. Because media relations is also becoming machine relations. 85% of comms leaders told us they know that earned media is training large language models like Tchaikapiti and Gemini. And 72% said that they've already changed or plan to change their strategy because of it. What does this look like? Pitching more digital first outlets, prioritizing formats that are LLM friendly, and monitoring how your brand shows up in AI search. So I'd love to see in the chat, has anyone already started monitoring how your brand shows up in LLMs? Is there a tool that you really like? I would love to hear your knowledge share here. But let me just quickly recap what's happening here. You probably already know this. AgenTex search, also called AEO, GEO, AIO, is the new SEO and it is powered by earned content. So when people ask tools like ChatGePt or Perplexity what product to buy or how a company stacks up against its competitors, these answers aren't primarily coming from SEO optimized sites. They're coming from press coverage, Reddit threads, Wikipedia pages, and comparison sites. That means your media hits, third party mentions, and community chatter now directly influence how your brand shows up in AI powered search. I want that to sink in for a moment. The most important societal shift in how people find information, that's coming from directly the work that we as comms people do. So every comm strategy also now needs to have an agentic search strategy. We can own a real seat at the leadership table in one of the most important conversations if we allow ourselves to deeply understand our company's agentic search presence and how we can influence it. But I have a chilling stat. This did not come from my research. This came from the correct data. Only 12% of PR pros plan to implement GEO strategies in their outreach for next year. Or 12% of pros plan to measure GEO in their outreach next year. Why are only one in 10 taking advantage of a huge opportunity to demonstrate business value? And I really hope that you'll be a part of that 12%. Final insight for today. Using AI to get better is only half of the equation. Have you told anyone? Because if AI is making your work sharper, but you're still scared to admit admit that you're using AI, you are missing out on the biggest opportunity. Let me say it again. Everyone, 100% of pros in the survey are using AI. The winners are going to be the ones who lean into it and, dare I say, do PR for their AI program. Let me show you the data. 93% of comms pros told us that AI helps them do their job better, but only 41% said it improves how others perceive their work. The missing piece is doing PR for your AI program. Why does this matter? Well, our data also showed that 71% of comms pros believe at least part of their job is at risk because of AI. And I'm a be real with you, besties. They're right. If you work at a computer, a 100% of you, some are part of your role is at risk. Mine too. So the worst thing you can do is hide your AI use and hope no one notices. That just makes you look replaceable. The best thing you can do is lean in, show your leadership, your peers, and your team exactly how you're using AI to make your work sharper, faster, and more strategic. Because the people who zig when everyone else zags, the one who embrace the tech, measure its impact, improve value, leadership is not gonna put you on the chopping block. They're actually gonna tap you for leadership. We live in an AI future. Your bosses want you to be native in this. So here's the template for how to do PR for your AI program. By using AI to task, we had impact. That meant business outcome. Let me give you a few examples. Instead of saying something like AI wrote this draft for me, you would say, AI helped us pressure test this narrative in half the time so we could sharpen the story in a line faster. Instead of saying, AI built our media list, you can say, AI narrowed our media list from 200 to 20 most relevant, giving me more time to personalize pitches and increase reporter response. Instead of saying AI saved us time, you can say AI gave us back five hours this week. We reinvested that time into building messaging frameworks and coaching the team. Alright. Those are my insights. I'm gonna hand it over to Shelby, and I'm looking forward to your questions. Thank you so much. Awesome. Okay. Well, let's like I mentioned, we're going to talk through now just strategizing this in a way, but also how does this impact the quoted side of things. I already saw some questions in there talking about quoted policy on, like, AI being forbidden. And that's the first, probably the best launching point to kind of set the stage for our stance. I mean, my personal stance, Quoted's stance on the use of AI. Quoted is Quoted is very understanding that we live in a world that is using AI. We greatly know that it is very, very important. It's going to help make things much, much easier. So I wanna be clear. We are not against the use of AI. We're against the misuse of AI. K? And we're gonna talk through what that misuse of AI looks like. Maybe setting the tone a little bit too is when we when we have these conversations on this on the the flip side, we're also having these conversations and training our journalists and our media users, and we're understanding the other side of our network and what they need. And the problem that our journalists are encountering, and we just had a full webinar last week, where journalists right now, they have the hardest time seeing who's real. How do I know who's real? How can I, vet verify individuals? How can I see if it's a false persona, or a real person? So that's really where we're trying to give everybody tools on both sides to help you stand out because reputationally, everything looks great. You're not making yourself look a little bit more generalized, a little bit more generic based on how you're putting AI content together or how you're summarizing and putting your content together versus those individuals who are just, not prompting their LLM models in the appropriate way and just putting in kind of falsified or fake content. So let's start first. And if you've been on any calls with me, you know that I love to talk about profile optimization. I think it is the most important thing that you can do with your quoted accounts. If you have not done it, again, get those profiles completed or configured. But let's talk about the do's and don'ts of using AI in building out these profiles and helping them to just engage in the network and helping them look really strong. So as I was looking at this, I was starting to think, like, what are ways that we can kind of bucket or give you some quick takeaways of things to keep in mind? And in my mind, the the bucket here is to use AI to enhance, use AI to pressure test, and use AI to position. Okay? So when we talk about enhancing, right, use it to improve clarity, polish, and maybe adapt your tone, shorten, lengthen, remove fluff, etcetera. So you can prompt your LLM to do those things. Use it to pressure test. So put your content in there and ask it to poke holes, find gaps, ask AI, what does this person actually do? When I put in when I put in that bio, is it indicative of what they actually do? Is it a little bit generic? Why are they qualified? What feels vague? What claims need further proof? Right? So if I'm saying this person's a leader in technology, that's a claim. So is the rest of that proof in the bio? And and the AI model is going to pull all of that detail out for you. And then positioning. So sharpening expertise, clarifying angles. Are we a bit too broad in our bios and in our profiles? How can we help narrow that down? So examples here, what are five topics that this expert can realistically own or speak on? What trending media angles most align with this profile? What headlines would this person logically appear in? And if those responses from your LLM don't align with where you're trying to position your profile at, that's probably a chance to kind of adapt and change that overall. Okay. So then on the flip side, for building out your profiles, do not use your LLM to exaggerate your expertise. Exaggerating is it's interesting. Right? Because it's not outright fabrication, but it is inflation to a point. So you might like, I previously was just telling you to enhance. And so enhance could also mean exaggerate at some point. So at what point have we enhanced it too far to where we've exaggerated then to a point? So asking those questions back to say, like, if AI is telling you a bigger story that aligns when you're pressure testing those credentials and when you're poking holes in it, if it's saying, okay. This expertise is much, much larger than where we actually wanna be, We've probably inflated or exaggerated a bit too far. So that will kind of help you pressure test and poke holes in that. So, overall, enhance strategically, pressure test really honestly with yourself on where we want that expertise to be, and then position that person really, really authentically with where their expertise is at. Okay. Now let's talk about pitching and the fun part where everybody's gonna come at me and say, Quoted has a very, very strict policy on this. You don't use AI at all. And that's true. If you're just plugging if you're going into Quoted and you're grabbing an opportunity and you're plugging that in, putting it in the chat, and then just copy pasting, and then plugging that right back to a journalist, unfortunately, that's going to look like AI. So that is going to be very, very difficult. Right? So how do we make you stand out for our journalist? How do you make your content look real, really quality as well? And that's what we wanna talk through here. So using use AI to clean but not rewrite your pitches. That's kind of the the main focus here. So when you just plug a pitch or your content, any type of content, what AI is going to do is automatically just rewrite that for you. So it's all about the prompts that you're giving that AI model just like Gab about. Right? So enter your response. Ask AI to fix your grammar, remove unnecessary words without changing your voice, without changing your meeting, or without rearranging your entire writing. Give it those specific prompts or say, what would you recommend? Don't change my writing, and just highlight the areas that you think I need to change. So, again, enhance, pressure test, and position. Give your LLM the appropriate prompts to rethink or give you areas that you might have missed or that don't make sense, but do not have it just completely rewrite that model or rewrite what you submitted over and start start it from scratch because it will read as if it is an AI written model unless we've coached it and trained it in a way that is going to sound very authentic. Use the LLM to also prioritize what you should be focusing on. And so, like, whether that's your you know, maybe you're looking at there's five opportunities for the day that are in quoted that are really applicable. My client probably doesn't have time. I don't really have time. So which ones should I focus on, and which ones are going to help put our best foot forward? So what we're saying here is use your LLM to say, tell me more about this journalist. Tell me which opportunities are best aligned with my client's expertise and that we should focus on first. Ask the LLM to identify what the reporter's really asking for. So whether that's data, opinion, trend analysis, ask help to structure your response accordingly, or also ask it to look at prior articles and then help you formulate how you should structure that response accordingly as well. Okay. And then the don't. I already mentioned this. Please do not just generate a response for the journalist, send it over to your client, get the approval, and then submit it, or vice versa. Don't even send it to the client and just submit it. I I always feel a bit like, like a school teacher scolding students when I talk about this sometimes. So I wanted to give a really real world example because like I mentioned, quoted sitting at a very interesting vantage point where we're just trying to help you put your best foot forward as it relates to what the journalists see, how they engage, and what they want to interact with in the network. So when we talk about this, it's so obvious when these pitches reflect as really heavy AI or not a lot of authentic, real quality content that you put together. So the the funny one that I have as, like, a reference or a story on is I was chatting with a journalist. We got on, and we were looking at her pitch submissions. And she was like, there's a lot of really good pitch submissions in here. And she's like, the interesting thing is these four all look identical to me. And she's like, wow. I really like them, and I think they you know, like, they're just kind of a little bit generic. And she's like, but when you look at them, they all have the same four bullet points. Right? So the prompt was give me four decoration tips for the bathroom that can help to elevate the space. That's all the prompt was that the journalist had. Right? So you have these interior designers and these individuals coming in and responding. And I think it was very simple, like, swap out the mirror, swap out the fixture, like, door handles, swap out put in a colorful rug. And then the interesting one that really made it stand out for all four of these, they all had they all had the bullet point that said, add some greenery. And it said something like, hang a sprig of eucalyptus on the showerhead. It will help give it a spa like experience. And I don't know about you guys. I've been in a lot of friends' bathrooms or bathrooms all over the place, and I've yet to see many eucalyptus sprigs on the showerhead. Chad GPT is recommending it apparently. And I just thought it was so funny because it was everybody had that add the greenery. It's a spa like experience and put the eucalyptus there. And that was that was the identifier in all of those. And so while it it maybe is a a good prompt, it was the like, that's not a natural prompt for most people. Right? And so even this morning, I was like, gosh. Is this a is this a applicable story? And I went. I put that same exact prompt into the LLM, and it gave me all four of those same bullet points. It said specifically to me, add a simple eucalyptus bundle on the showerhead, and it can help make the space feel spot like. So it's just gonna be a different variation of that same bullet point. And so that's really where make yourself stand out, diversify those bullet points a little bit. Even when you are putting in maybe a bullet point that might be common knowledge, help put in some real authentic detail there because that's what your LLM cannot do until you've trained and coached it accordingly so that it doesn't just look like it's scraping everything from the Internet of what what one interior designer said at some point, and now it's giving that feedback to everybody else as they plug those questions in. So I'm gonna pause there. I wanna turn it over to our q and a. I know we have quite a few questions in the chat here. Liz, it might be helpful for us if you come in and maybe give us a little or we might be able to do it here just kinda going through because I think we have quite a few questions. Maybe give us some to focus. on first. Actually, the very first one is what's your take on some reporters, media outlets that are apparently vetting pitches for AI use? For example, what if I write a pitch and then use AI to improve or refine it and then copy and paste it into an an email, it could appear as though the entire pitch was written by AI. Yeah. I think that's where using those those prompts, like, don't please don't rewrite this whole thing. Please point me to my grammar my grammar errors. Please poke holes in this. Please help me enhance it or show me areas where I can enhance it. But please don't rewrite or change my language, change my tone. I think the the biggest thing here is when a journalist is vetting and verifying pitches, they have no way of knowing your intentions. Right? And so I have the best intentions that I gave my four bullet points that are great, and I think they're wonderful. And what I'm going to do is I'm gonna save some time, and I'm gonna put those four bullet points, and I'm gonna have Chatuchipiti write the answer for me and then plug that in. And that is where that will reflect as though it's AI. Even though the bullet points might be my own original thought, the writing itself is not original original to to me. Me. And so a journalist has no way of knowing, did Chatuchipiti write this? This is a fake persona potentially. This is a person who maybe isn't even an expert in this space. I don't have the time and the bandwidth to sit and verify and go back and forth. So I'm probably gonna pass along and go to the next one so that I can get a better feel for what that looks like. So I would say it's more about the the lack of understanding how you created it. They don't get to see the back end and how you created that content or how you put that together. So if there's a flag on it, it's like, ugh. I would rather not risk it and move along to the next one that I know doesn't appear as AI. So use those prompts. Make sure that's not rewriting it and it's not reflecting as AI, but, still utilize it. So, like I said, poke holes. Do all the other things. Fix your grammar. That's what I use it for mainly. It's more of a grammar checker for me half the time because that's what I'm the worst at. So kinda utilizing it in that space. Can I, like, add a philosophical addition to this? So. do you have good insights? None of this should matter. I understand AI flagging and AI detection, and I do not know the technology well enough. But, like, Jackie asked this question. Jackie's an off the record member. Jackie was on a session I did yesterday for our membership where I talked about this concept of you've heard of garbage in garbage out, but I'm an optimist, so I like to flip it on my head with on its head, and I call it diamonds in diamonds out. If you're working with a brand or an executive who doesn't have an original thought in their life, you are not gonna stand out. These are not gonna be good pitches. If you put your pitch into an AI or put the question a journalist has put into AI, what is that answer? Your answer should be different. It should be spicy. It should be counterintuitive. It should be slightly controversial. This is the type of content that regardless of if you had the original thought or not, when journalists say this sounds like AI, it probably sounds basic. And you are going to need to encourage your clients to have better thoughts, better diamonds in. Thanks. I. love that. The next one is, are journalists using tools to block AI generated pitches? And if so, what tools? Great question, but I don't know if we can answer it. I see Scott Steinhardt on here. Hi, Scott. Scott's also an off the record member. He runs a or he runs comms for a deep fake authentication company, and they have built, like, internal systems. So I don't know, Scott. Maybe you can do a LinkedIn post for us soon about the different tools that you might suspect journalists are using to detect AI. Yeah. There's a it's hard. Like, none of them are gonna be perfect. Right? And I think that's the the biggest thing. You know, I see I see, like, Jennifer Thomas is on here. I know she's also a writer. She's kind of utilizing quoted from both sides. Quoted a year ago, we or or actually, it was a year ago, we integrated in AI detection tools where you can check a pitch. It's on both sides. But we as of the last six months, we switched over to Pangram. We found that Pangram is it's the most reliable. So this is not me saying like, hey. Pangram's never wrong because these tools, like I said, they're not perfect, especially in this AI world that's ever changing. You know, talk to me in six months, and Pangram might be completely behind, and we're continuing to try to keep up on that research. What we found, though, was when we were asking journalists what types of tools that they were using, they were like, whatever's free and whatever's available for me to use. And so sometimes those free tools are the ones that we were concerned with because it could be giving those false positives or those bad reads. So we decided to integrate Pangram into both sides because we feel really confident. It has the least amount of false positives of anything else that's available right now as far as, like, the AI detection and the test there. So that's the one that we know in Quoted that they are using. But, yeah, there's a there's a lot of different tools. There's a lot of different companies coming out. It'll be really interesting over the coming year, I think, just to see what continues to change. I even yesterday, this is a I mean, shameless plug for them, I guess. We chatted with a company called Okay Human, and they're creating digital signatures for writing. So that that way, when you are submitting, when you write it, it will add this little digital signature that is specific to that person. So I have I created my own digital signature yesterday. The company is free. You can sign up. And what it will do is anytime you you write that content, you can edit. You can change it or whatever, but you can have that digital signature that shows that it was genuine written content by you. So I think there'll be many companies like this coming out that are just trying to show us who's real. Because like like everything, there's just really bad actors in the space that kind of ruin these tools that are really, really helpful and useful for everybody where we then have to verify, like, who's who's the real people in all of this this swap that's being maybe submitted. And I think that's where the relations word in media relations is super important. Because I think from the journalist perspective, like, there is nothing more shameful than if they publish an article with AI generated companies, data, book lists. We've seen this before. It's probably hellish for them. So as they're working with a platform like Quoted, they're like, I like, this all has to be real. This has to be a real person with these real credentials and real insights. So ask yourself, what can I do to prove to them I'm a real person? Maybe after. you send in I'm I'm I'm actually not sure if you can say, like, the journalist's name and every quoted thing, but, like, maybe you also send them a message on LinkedIn. Or maybe you include, like, a picture of yourself in the thing. Or maybe you say, like, here's other coverage that I that my expert has done on this topic. Something to validate and, like, add more credibility to who you are and continue to build that relationship. Yeah. Exactly. The next question combines two questions. It's about building media lists using AI. So Marjorie asked, we want to build a media list to include outlets that AI relies on most. How can we do this? And then Kurse asks, regarding using AI to vet a media list, how do you do this effectively when many media outlets are gated or block AI from crawling their site? I can hit the first one, which is how do you build a media list that includes publications that are more likely to be cited by AI? One key thing is look at the publications that have partnerships. So, like, Business Insider and OpenAI have a partnership. So, you know, Business Insider is a top cited publication for the AIs. Muckrak also releases data on this all the time, and they will actually break it up by industry. And I think what you will find is two big surprising insights. Niche industry trade publications and the are are are very highly cited in these as well as, like, kind of, like, listicle or, like, product review sites that you maybe, like, weren't considering before. Like, g two, if you're a tech platform, is, like, highly cited by these these bots. So I would make sure to look at that. And then the last thing that you can do, just like be scrappy. Right? Do your own searches in a couple of LLMs. Ask them to cite their sources, see what publications are coming up, and then that can help you target your list. But here's the deal. It's not much different than what you all have been doing for the past decade or two. It's finding the right audience. And if your audience is supernation specific, then supernation specific outlets are going to carry more weight. You can get something in the New York Times, but there is no guarantee that your buyer of widget is reading that article. But when you do widget trade magazine, you know for sure the readers are doing it. Same stuff, y'all different audience. Shelby, I don't have anything to add as far as using AI to build media list. That's not something that I it's not really even probably something I would recommend. Your your taste and judgment on who is the right person for this story and who is their audience, that that probably reign supreme. And that would be an activity that if I was still actively pitching, I would feel I mean, I would be what if I sent something to a client and was a it was a hallucinated journalist or publication? Like, I would be horrified. Yeah. Yeah. Yeah. I I would say the same. I mean, I don't have a ton of experience in doing this, and I think it it's hard because I think just media databases in general and building those media list is it's quite difficult just because, as you guys know, the the journalism market is changing every single day. People are being laid off. People are moving publications regularly. Freelancers are writing for multiple publications. So contact details, all of that is pretty elusive. Quoted Quoted has a free media database that we've included. This isn't a plug for you guys to go and use the Quoted media database candidly. This is more so to use it as a cross reference, and that's where we wanted to make this information to use almost, like, as an anchor because it's so often when you you pull a media list, you send things out and you get bounce backs and you get, like, all these, like, this information's not up to date. And so the more areas where you can cross reference, like, whether I'm looking at LinkedIn and Muckrack and Cision and Agility, and then I'm logging in to Quoted and I'm cross referencing. Like, ideally, I'm finding at some point the right contact details or I'm filtering this list down. So I would say even using AI to kind of collate all that data that you're getting from those places to show me the discrepancies or show me like, hey. These ones were wrong. Can you find me the correct contact details elsewhere? Because I think that's that's probably the more challenging aspect or at least what we hear about it quoted, the most challenging aspect is, like, how the heck do I find this information? Okay. Sorry. I was looking at some of the questions in chat, and I'm gonna jump in and just say there's a question of there are many comms professionals here. Do you typically work with midsize to larger companies, or do smaller companies also engage your services gap? Me? Mhmm. Oh, we inside off the record, I would say we are primarily around the director level, primarily ten years of experience. But I think that's great. And prime we're primarily in house too, but we do have consultants and agencies sizes all over the place. We have folks who just work for themselves as a consultant and folks who work for the number one, two, and three companies in the world. So it it runs the gamut. If you care about delivering business value and you want to grow in your career, we got stuff for you. Thank you. Moving back to AI. What are some of your useful AI tools? A lot of people wanna know. Yeah. Let's go. Okay. The tool that your company allows you to use, first off. But if they don't and you maybe, like, accidentally purchase a separate computer so you can access any tool, not recommending that, but I also don't work for your company. I my strongest writing partner is Claude. Claude does not edit my content very much at all. It does kinda what Shelby's talking about about, like a grammar checker. I have been noticing some parts of that platform just, like, depreciating a bit on its quality for me. So I'm about to embark on an experiment with Google Gemini Gems to see if I can have some better results out of that. Now the one that I would I'm actually gonna grab my little my little share link on this one. Y'all gotta check out Lovable. It is so cool. It is a vibe coding app. It is like prompt to website, prompt to app. It's a very cool platform. Just play around with it. If you use this, you will get 10 free credits if you sign up from that. So that's a that's a great platform as well. But I just think, like, use the one your company approves. If I say Claude, you can do it for Copilot. If I say ChatuchPety, you can do it in Perplexity. Like, they all kinda have the same general bone structure. It's just a matter of which one is your preference and which one have you formed that parasocial relationship with your AI tool, or is that just me? Yeah. I love it. No. I'm also I'm mainly using Claude as well because I found that it it doesn't rewrite my my stuff as heavily, or I don't have to prompt it as much to not rewrite it and kind of work backwards. I'm not gosh. I'm trying to think outside of what you said. I've been using speak.ai, which is really great for those of you working with clients. Oftentimes, you might send them questions. They're on the move. They're running companies. They're very, very busy. What it allows you to do is it sends them a link. They can speak in to the you know, just they record it, voice note in. It's like a voice note text, essentially, is kind of what it is. And then it will come in, and it will summarize it. But it will summarize it in a way that it does not change the content itself, and then it will help you highlight areas where it, you know, where it captures, like, he said this in a run on sentence, so we kind of need to tweak or adjust this here. So I found that that's really, really helpful for those clients that are maybe not writers, don't have time to be writers, don't wanna sit down and type something out for you. They can kind of just quickly speak into that. It kicks back over to you, and then you can plug it into everything else that you're using. Two others that I'll name. One is I don't use this one. It's called granola. It will record all of your meetings. You can chat back with it to kinda get insights from those meetings. I do not know if Granolah does two way consent recording, so you're gonna have to make your own ethical decisions about that. And then the second one that I love and is such a kinda untalked about AI tool is Google's Notebook LM. You can get access to this with your personal Gmail account. And what I love about NotebookLM is that unlike AI models that have a hallucination problem, NotebookLM will not hallucinate. You can upload literally 20,000 pages of content and ask it questions specific to that. So that the OG use case for Notebook LM is you upload a company's financial statements or their, like, earnings call, and you're like, literally, what is this gobbledygook? I have no idea what's going on. And it will give you a summary or it can even make you a podcast of that. But for me and, like, what I do is I do hundreds of interviews for all the content that I create, and then I have my own insights. And I put all that in Google LM, and that's where I refine. How do I take a hundred hours of content and bring it into a one hour training session? Google LM is a great partner on that. Notebook LM. I'm calling it the wrong thing. Y'all know what I mean. Wait to look at that one. Ryan. Hi, Ryan. Asks, in addition to earned media, if you're getting press releases published, is that also an effective strategy for building up GEO? Okay, friends. Let's talk about this one for a second. There has been this press release renaissance that makes me want to, like, go scream at a wall and, like, punch bricks. So let's just change the language of what's happening here. You know, I know, we all know. Unless you are a one top 1% name brand company or have, like, the best data in the world, your press release is not going to get you media coverage. You all know this. That's why you have jobs. If that was the case, you wouldn't have jobs. We wouldn't have jobs. Right? Press release does not equal media coverage. Let us change the name of what is happening, and let's call them AI notices. So the press release format is having a resurgence because the AI models, for reasons that I don't agree with, but I will explain, trust the press release as a factual first party data. And here's why. You've all put out a press release. Right? You know, you go into the system and you type it in, and then you have to get that call from the editor and you have to, like, approve the headline and who it's from and your name's attached to it and God forbid your personal phone number. And so everyone calls you all the time. So that process of putting it into a system and a human editor calling and confirming from a human inside the company to confirm this information comes from the company, the LLMs, like, salivate off that. They think that's, like, the most, like, juicy, like, reliable information. So that is why press releases AI notices are having a resurgence. So from what I understand, I'm sure they're all getting in on this game, but the OGs on it was Globe Newswire. You use their online only. It's, like, $250 a pop that you GEO optimize it using best practices. Or if you're not the record member, you can use my tool to GEO optimize it, and then you send it out. But, obviously, you still have to do media pitching. This is just a part of that GEO strategy, that machine relations part. This is not media relations. Okay. We can all agree. Alright. I see a lot of stuff coming in. Hopefully, you guys agree with me. If not, I'm excited to read the comments. How about. any insights on how to handle clients who are too into AI and want to use it as the only and primary source and tool for nearly everything from podcasts, editorials, etcetera, essentially concerns about ethics and protecting the brand's credibility, and that comes from Amy. Gosh. I mean, I think this goes back to that that diamonds in, diamonds out. Right? Like, it's I think when you I I think there's so many like, you can even think about, like, when you're scrolling your Instagram at night or whatever, it's like you can see these companies that just have AI generated ads and AI generated TikTok feeds or everything else. And I think it just it doesn't look good reputationally. It looks very generic and very basic. So maybe as, like, a supplement to, like, maintain the level of content that they're hoping to put out with low resources, but I think that it needs to be supplemented with, like, real, really genuine authentic content, or else it just it just looks really generic. And I like I said, on the the journalist side of things, so many journalists, they're like, I can smell it from a mile away. I can tell exactly what this was. And when you're submitting on their behalf, like, if you're working with other clients and you're submitting to that journalist, but you wanna work with that journalist again, like, reputationally, like, I don't wanna submit that. Like, if I know this is generic, I know it's not my best work. I don't wanna hurt my reputation with that journalist long term for my other clients because that's something that just looks like I don't probably don't wanna work with this person again just, like, kind of considering what they're willing to send to me. Yeah. I I like, my favorite thing that's happening is that people would be like, full disclosure, I made this image with AI. And I'm like, really? Like, I knew. Like, it's obvious. Yeah. Right? Like, Yes. it's it's clip art. It's like you can tell. It's like it's it's lazy work. Right? And so I actually do think, like, designers with taste and judgment will continue to have great careers because, like, I got a email yesterday for spring break camp for my daughter, and each one, like, had a little image. And I was like, AI generated AI gen which, okay. Fine. Like, I'm fine if her little camp wants to, like, do AI generated images, but, like, I know. I know. You did not you did not fool me. So I actually believe that not using AI is a brand differentiator. And you're seeing this with companies like Aerie, which is owned by American Eagle. That's a different story. Let's talk about the Aerie brand. It's like very real human. They were like one of the early people did not use Photoshop on their models, and they they show models of all different body types and sizes and heights and standard definitions of beauty. They they challenge that. And now their thing is like, do not use any AI generation with our models. So that really leans into what the brand is. Now if your brand is an AI company and your OpenAI, they actually even got some flack that they didn't use Sora or AI to build their commercials. You know? So it's like, oh, are your tools actually not that great? So it just kinda depends on, like, what the expectation is of your brand. Okay. So how do you train an LLM to talk about you and your agency or your client and their leadership, does that affect others who are researching you? Do we think this is a GEO optimization question? I I think it's more. about doing those projects and feeding feeding your LLM, Claude, whoever, data about your your company or your clients, creating a voice, etcetera. Yeah. Yeah. So to Scott's comment here, every LLM tool has the version of a project or chat GBT, custom GBT, or Google Gemini Gem, or a Copilot agent. And, usually, those have an ability to upload documents. In Google's case, you can dynamically link to an Google Doc. So I interviewed someone named Neil Ripley who works at Premunitive, and he talks about how he uses Gemini gems. And one thing linked in there is his master messaging document. So every time that messaging updates, the gem is updating too. So you just have to give it tons of diamonds, tons of high quality previous communication samples, clear direction. In mine, I have things like, don't use the word actually. Like, come on. It's always used. Or, like, the honestly, question mark, the real thing is it's like, I don't know where I give you valley girl vibes, but, like, I am not here for it. So you can just, like, kinda add in those rules and stuff. And when in doubt, just ask your bot. You're like, hey, yo, bro. How do I do this? And they'll let you know. Should PR pros be careful or particular about what they put into their LLMs about their clients? Should they Ryan asks, should they only use this information in an incognito chat or use fake names? You know, it's he and then he goes on. Y'all gotta pay $25 a month. Like, you gotta get a you gotta get a paid subscription to these tools or else like, you're not gonna get any sort of meaningfully output. When you're a paid version, you can ask it not to train on your data. When you are an enterprise, you definitely aren't getting trained on the data. I wouldn't recommend against putting PII or, like, trade secret or, like, classified information into LLMs. It just depends on if you trust if they're actually going to not train like you say it is. Back in the day, I would put in, like, fake names. I would always, like, control c and say my company was called Acmeco because I was, like, a little afraid of it. Now it's like, oh, this is gonna be public in, like, a week anyway. Like, meh. Who cares? Like, let's just let's just have it go. So just I think I think just be smart. Thank you. Michelle asks, from your webinar with journalists, how many are using Synth ID to check for watermarks and copy? Even if a pitch is refined in AI, they might input a watermark, which then a journalist will not use even though not completely generated by AI. And she Michelle, that's a PS. PR pros are not allowed to join Synth ID. Interesting. Yeah. I mean, we didn't ask that question, so I don't have an answer on how many of them are using it. I think that's where you know, like, brought up Okay Human, and I think that's something similar to what they're trying to do. Yeah. So it's I think that like I mentioned, it'll be it'll be interesting to see how this develops and changes over the coming year. I know on the quoted side of things, there's plans. So when you are submitting that content in, there's ways to help kind of showcase that it's verified for lack of a better word there. Right? Like, it's whether that's your profile, we're trying to partner with Clear so that you can Clear verify those profiles to show which ones are real, whether that's your your pitch itself, of putting that stamp on it to showcase that that was really your content, or if it's uploaded image, if it's uploaded video, even now we have to worry about. So, yeah, it'll be interesting to see how it changes. Again, we haven't asked the journalists that. It'd be interesting if we do any type of survey of them soon to definitely keep in mind on if they're looking for it. Thanks. We're. almost at time. We've got about just under four minutes. I thought maybe you two would like to share some final thoughts and next next month's webinar, and I'll go away. Awesome. Yeah. I'll let you, share some final thoughts if you want, and then I can close her out. Yeah. Absolutely. So, you know, you I think the number one thing that I would recommend for everyone on this call is to block out time this and next week to experiment with some of what you've learned today. Use AI to automate the stuff that is not a strategic value of your big, beautiful brain, and then talk about how you're using AI to make that big, beautiful brain get to spend more time on strategic value. But you have to do it. Right? Like, webinars like this are good. Learning, like, on things like this are good. I'm gonna send you more stuff so you can learn even deeper. Like, you just gotta kinda have to do it. A quick example for me. One of my friends in comms recently became the global head of comms at Lovable, which is a vibe coding app, and we've explained what vibe coding is. It's basically like you can create stuff without writing lines of code. And I was like, oh, what should I, like, use Lovable for? She's like, girl, you just gotta, like, get in and use it. Now I'm obsessed, and I have created stuff for me, my business, my members, my daughter's theater company, my daughter's school. Like, I am a huge stan. Right? Because because I started. Because I did something. Because I acted. I didn't. give up when it was maybe not the best first draft. I questioned myself. I said, how can I feed this Better Diamonds? And I saw what is possible. And now there's just so much more time and opportunity unlocked for me to continue to do what I love, which is deliver value for comms people. So just try. Just see what happens and do something. I love it. Yeah. It's certainly like, I think you mentioned, like, continuing to just, like, educate yourself in the space. I think it's it's going to be ever changing. There's no way that we can give a a straightforward answer on a lot of what this is going to look like. It's everyone's learning and kind of adjusting and adapting. So the more that we can kind of help to provide you guys resources, we we will do. But also sharing with your network and trying these tools and and figuring them out as well. We do these webinars every single month. I hope that we can have you on again soon because I think that there's definitely an extra layer that we can dig into. This is kind of like, as we talked about, kind of our basic one zero one on how to get started, but I think we can move more into that upper intermediate advanced area that we also wanna dig into as well. So we would love to have you back, but, we also wanna make sure that we're putting out content that's really applicable to everybody. So, as I mentioned, we do these every single month. I just shared the academy link in the chat as well. So we upload these events. If you're not able to attend, as long as you register, we'll email you the recording. You can also see the recordings in that academy. We just officially, locked in Jeanie Dietrich for next month's webinar. We are going to talk about earned media and what you do once you get it. And so so many people, I probably get this question 20 times a week. Hey, Shelby. I just pitched and quoted. Now what? And so we're gonna we're gonna really talk and unpack a lot of that. What do you do with all that content once you get it, once you have those placements? How do you track down content that you might have been featured and then you're not sure of? So we're gonna unpack all of that next month. So that will be on March 26. We will have this recording out. We will continue to answer any questions that you guys have. If you wanna jump into the quoted chat, you're always welcome to jump in there to our support team, and we're happy to pass those along and get further questions back to you, but, happy to help in any way. Okay? Thank you, everyone. Thanks for having me. Bye. Bye.