No more excuses. No more waiting to see how things play out. AI has moved past the experimental phase, and if you’re still treating it like a nice-to-have rather than a fundamental shift in how your agency operates, you’re already falling behind.

In this episode, Chip comes out swinging with a wake-up call for the agency community: the ground is shifting faster than most are willing to admit, and the window for meaningful adaptation is closing. Gini backs him up with examples of how AI has progressed from an intern-level tool to something that can genuinely replace mid-level work—if agencies don’t evolve what they’re selling.

They dig into the practical reality of training AI tools to work like team members, not just one-off prompt machines. Chip explains how he uses different platforms for different strengths—Claude for writing, Gemini for competitive intelligence, Perplexity for research, and ChatGPT as his strategic baseline. Gini shares how her 12-year-old daughter creates entire anime worlds through conversation with AI, demonstrating the power of treating these tools as collaborators rather than search engines.

The conversation covers what clients actually want to pay for in 2026 (hint: it’s not social posts and press releases), how to build AI agents trained on your specific expertise, and why the process of training AI forces valuable clarity about your business. They emphasize that this isn’t about slapping the “AI-powered” label on your services—it’s about fundamentally rethinking what value you deliver and how you deliver it.

If you’ve been sitting on the sidelines waiting for the AI dust to settle, this episode is your warning: there is no settling. There’s only evolution or extinction.

Train one AI tool this week like you’d train an employee. Pick the platform you use most (ChatGPT, Claude, or Gemini) and spend 30 minutes having an actual conversation with it about your preferences—tone, structure, what you hate (like emojis), and what outcomes you need. Feed it examples of your best work and tell it explicitly when outputs miss the mark and why. The tool won’t improve with one-shot prompts; it needs training just like a new hire.

Map what clients will actually pay for in 2026. Block one hour to list every service you currently bill for, then honestly assess which ones AI can now handle at a competent level. Don’t lie to yourself—if ChatGPT can draft solid social posts or press releases after reviewing past examples, that’s table stakes now. Identify what remains valuable: strategy, teaching clients to use these tools, implementing new processes, or solving problems AI can’t touch. This clarity will drive every business decision you make this year.

Test AI on something personal before rolling it to client work. If you or your team are intimidated by AI, start with meal planning, fitness routines, managing schedules, or drafting birthday card messages. Use it for something low-stakes where you can experiment with conversation-style prompting without pressure. Once you see how it responds to feedback and training in a personal context, you’ll understand how to apply the same approach to agency work.

The following is a computer-generated transcript. Please listen to the audio to confirm accuracy.

Chip Griffin: Hello, and welcome to another episode of the Agency Leadership Podcast. I’m Chip Griffin.

Gini Dietrich: And I’m Gini Dietrich.

Chip Griffin: And Gini, you know, we started the new year off on a note where we weren’t gonna yell at our audience, but I feel like it, it’s time to yell at our audience again. I’ve taken too much time off from being Mr. Nice guy.

Gini Dietrich: Okay, well this shall be interesting. I can’t wait.

Chip Griffin: I, and this is, it’s partly for our audience, but it’s really for the overall agency community, particularly PR and marketing, PR and communications generally, even outside the agency world. I’m just, I’ve become kind of wound up lately because I think that the industry as a whole, and perhaps even some of our listeners are not acting swiftly enough to understand just how much the ground is shifting beneath them.

Gini Dietrich: Yep.

Chip Griffin: And how much serious evolution needs to take place. Really over the next year. I mean, I don’t think, I don’t think we’re on a long-term horizon here. I think that too many have waited to change too long in many ways, and AI is now becoming sort of the, the real trigger point for it, but it’s bigger than that.

I think a lot of the, the PR space in general has lagged behind a lot of what’s going on in the business community, and AI is just the fist to the face that’s, that’s gonna separate out the people who are gonna survive.

Gini Dietrich: The fist to the face. Wow. All right, then.

Chip Griffin: I told you I was a little wound up on this one, so,

Gini Dietrich: okay.

So everybody’s gonna be punched in the face. Got it. Okay.

Chip Griffin: If that’s what it takes to wake up and pay attention.

Gini Dietrich: Yeah, no, I, yeah, I totally agree with you. And, you know, I have been gungho on AI for going on four years now. And it’s, it’s my second love for sure. But it is time to pay attention to how it is changing things and what it’s going to do to your business, to your teams, to how you deliver work, all those things.

Chip Griffin: I mean, look, a lot of the PR world has been focused in recent years on figuring out how to keep their head above water and survive, and hang on to the old ways of doing things. And this predates the explosion of AI in recent years.

Gini Dietrich: Yeah.

Chip Griffin: But, what the explosion of AI has done is really, it has drawn the attention of particularly clients to the issue.

It has drawn the attention of employees. It, and it is still being ignored. And I think we’ve hit that point where we can no longer ignore it. I think we’re at the point with a lot of these AI tools where they are now both accessible and reliable enough that there’s no reason not to accelerate your pace of change using AI as a tool to get there.

And we’ve talked about this before, and I, and I’m not changing my point of view, AI is not the end in itself. The AI is just a way to get there. So don’t mistake what I’m saying here for saying that, you know, you just need to adopt AI for the sake of AI. You still need to find problems to solve first and AI will help you on a lot of them, but you need to be finding those problems.

You need to be thinking ahead to what do clients really want from you? What is going to help them to get the results they’re looking for? It can’t be about how do I use AI to make myself a little bit more efficient in what I’m currently doing. Because everything is changing. And we need to be on top of that.

Gini Dietrich: I read an article on LinkedIn probably in November, and I’ll see if I can find the link to include in show notes. But it, it was from a chief marketing officer at a Fortune 10 company, and what he said was this: if I were an agency wanting to work with clients in 2026, here are five things I would do. And I can’t remember all of them, but one of them was teach organizations, teach marketing and comms teams how to use AI to be more effective. Implement your process, whatever it happens to be. Because we are no longer relying on our agencies to do the work. We are relying on agencies to teach us what’s coming ’cause we don’t have the time. And that has stuck in my head because I think that’s right.

I think that. Yeah, sure, agencies will always, or big companies, will always need arm extra arms and legs to do the work, but that’s not the work that most of us want to be doing. Right?

We don’t wanna be writing the social posts and the news releases. We wanna be part of the strategic conversation. We wanna be part of the of helping to move an organization forward. And if we can do that by teaching our clients how to use AI to be more effective, to be more productive, to accelerate their work, and I know everybody’s worried it’s going to replace me, it’s going to, it’s going to reduce our number, our billable hours, whatever happens to be.

I think there’s a huge opportunity here for you to reframe how you’re helping clients and using AI to be able to do that.

Chip Griffin: Yeah, but I would be very direct with listeners. If you do not change, it will replace you. It will take away.

Gini Dietrich: That’s fair. That’s totally fair.

Chip Griffin: Your revenue.

Gini Dietrich: Yes, it will. I totally agree with you.

Yeah.

Chip Griffin: So, you know when we say that you know that AI is not gonna destroy your agency, that’s only if you evolve.

Gini Dietrich: That’s fair.

Chip Griffin: If you keep doing the same thing that you’re doing today, it absolutely will destroy you. I don’t care whether you’re an employee or a business or whatever, if you are an employee and you think that AI isn’t gonna take your job in a year, it is If you don’t evolve, that’s and figure out how to use it for yourself.

Gini Dietrich: Yep. That’s totally fair.

Chip Griffin: And we need, everybody who’s listening needs to wake up to that fact.

It requires a huge mindset shift.

Gini Dietrich: Yes. Because AI can write your news releases, it can write your social posts. It can do all of that stuff that

Chip Griffin: not only can,

Gini Dietrich: we don’t wanna do anyway,

Chip Griffin: It should. Yeah.

Because it has evolved enough in the last year that the quality is there now. I used to describe AI as an intern. It is moved beyond the intern stage. Yep. It is at a minimum a junior employee, and if you train it well for your organization, it can be even a mid-level employee or perhaps even in some cases more than that.

But this training piece is important too, because part of the problem that a lot of people run into in my experience is they, they hop onto the AI tool and they just say, Hey, write this press release on this subject. And I look at it, oh, this is rubbish. It still requires a lot of work. You know what? It absolutely does.

The same thing would happen if you hired an employee off the street who knew nothing about you and your clients, and you said, write me a press release. The result would probably be pretty similar to what the AI came up with.

Gini Dietrich: Yes.

Chip Griffin: But once that employee starts writing more press releases and you start telling ’em, this is the tone of voice we use, this is the style we use, these are the facts we use.

You feed more information into it. You explain your preferences. When you’re using these AI tools, you need to just be direct with it. Don’t accept the first response. Explain as you would with an employee what you want done differently. If you do that, it will tailor the outcomes. Even simple stuff. Like I’ve told them, stop showing me damn emojis.

I don’t wanna see an emoji in any response because I think it’s wildly unprofessional and I hate them.

Gini Dietrich: Yep.

Chip Griffin: So guess what? I don’t see them anymore.

Gini Dietrich: Yep.

Chip Griffin: I’ve asked it to tighten up the spacing on it so that I can see more on a screen. It does that. And that’s even before you start telling it, you know, this is the structure of a paragraph that I like.

You start feeding in information. I’ve fed in a thousand articles and transcripts and that sort of stuff into the platforms. It now can speak like me reliably to the point where I don’t know if what it’s giving me is a quote from something I’ve written before or original text that it’s come up with that just speaks so clearly in my voice.

Gini Dietrich: I love that it will say, it will give, usually gives me three options. One is like strategic leadership, like C-level blurbs. That with Gini-isms or like smart, funny, witty blurbs. And then I can decide, and usually what I do is I take a combination of the three, but it has gotten to the point where if it actually calls it Gini-isms, that like it knows how I talk, it knows how I write, it knows how I coach, it does it knows all of those things.

And it has created an opportunity for me to say, yeah, this probably, we probably shouldn’t have some Gini-isms in this ’cause it’s really professional. Or, we can include more because it’s more me talking to a screen or whatever happens to be. So it’s gotten to that point. It’s, when you train it, it’s very, very good.

Chip Griffin: Well, and you can even tailor those recommendations. So one of the things that, that I’ve told it is it’s fine to give me multiple options, but give me your recommendation.

Gini Dietrich: Yep.

Chip Griffin: And when you do that, don’t give me a whole lot of backup on the alternative. So spend your time explaining why you’re making the recommendation.

That’s fine. But then, you know, if it’s, let’s say it’s a title or something like that, you know, give me three or four other options, but it, by default, it tends to explain those three or other four other options. And so now you’re dealing with like a 10 page response,

Gini Dietrich: right?

Chip Griffin: For what should be something pretty simple.

Gini Dietrich: Yes.

Chip Griffin: So. I, part of my instructions to my tools are, don’t do that. Give me the alternatives, but just, you know, bullet point them. If I want more information, I’ll ask for it, but it allows it to work more the way that I want it to. And so we all need to do that. We also need to be looking at these tools and understanding that there’s no one size fits all solution.

I have people say, well, should I, you know, should I use Claude or Chat GPT or Gemini? The answer is yes.

Gini Dietrich: Yes to all of them.

Chip Griffin: But they all serve different purposes. Yep. Just like you have different employees who serve different roles, these tools excel in different areas. I mean, Claude is fantastic at writing.

I mean, to me, Claude is my head of writing because it can just absolutely nail it, but there’s a lot of things that it doesn’t do quite as well.

Gini Dietrich: That’s right. Yep.

Chip Griffin: And then I look at something like Gemini, and I love what Gemini does in terms of inferring things from research. So it’s more willing to go out on a limb, and kind of read between the lines of things that it finds to come back with, particularly for competitive intelligence or things like that. You know, deep research.

Whereas Perplexity is very good for research where you really wanna make sure it’s accurate and you really wanna be able to cite all the sources, but it will not go out on a limb. So understanding what the strengths of each platform are is useful. And then there’s Chat GPT, which is sort of my, you know, my default choice for just basic stuff, strategy, et cetera.

But I’ve also told it, tell me when I should go somewhere else. And so it’s good. It’ll say You should hand this off to Claude now.

Gini Dietrich: I love that.

Chip Griffin: Because we’ve, I’ve had an actual conversation with Chat GPT about my stack and, and what I think of it and I bounced things around and, you know, refined it.

So now it knows how I want to handle certain things. And so it will stop at a certain point and say, now it’s time for you to go here. And that’s really helpful.

Gini Dietrich: I love that. I do not do that. I usually move between, but I haven’t had it recommend when to move it. That’s…

Chip Griffin: Yeah. I mean, but it could, because it won’t generally by default tell you to do that.

But if you, if you explain what you have access to and what you want to use it for, it will tell you when is the right time, and sometimes I’ll pause and say, are you really the right one for this? Or should I be using one or the others? And they’ll say, no. Good point. You know, you should use this one instead for this particular task.

Gini Dietrich: I love that.

Chip Griffin: And it’s great. I mean, and I’ll, I’ll bring things back and forth like, so when I’m creating a piece of content, I’ll often, you know, ask more of the strategy piece from Chat GPT, because I’ve put more of the strategy stuff into there. Then I’ll go over to Claude to write it, but then I’ll bring it back for feedback.

Now the next level is then to automate this with agents with n8n and those kinds of things. And, and so, you know, I’ll play with those things too. But for now, even doing it manually is a huge time saver,

Gini Dietrich: huge time saver,

Chip Griffin: and still ends up with really high quality content. It’s not, people talk about how AI is helping put out rubbish.

And that’s because people are doing it without training.

Gini Dietrich: Correct.

Chip Griffin: You need to think through how you use these tools to get the results that your clients are looking for and the results that you need as a business. And this is where people are falling down, and this is where a lot more effort needs to go into it.

If you want to not just survive but thrive.

Gini Dietrich: Yeah, I totally agree with you. And you know, it’s funny ’cause when somebody says to me, oh, I just can’t get it to output what I need, and I’m like, user error. I usually say that because that’s exactly what it is, is you haven’t taken the time to train it. I, and you have to, I, you said earlier, you talk to it like it’s an employee.

I do the same thing. Talk to it through like, okay, this isn’t quite right and here’s why. Think about this, this, this, and this. We also need to consider these things. And then it goes. Oh, okay. Goes into thinking mode and then it, it outputs pretty close to, but you have to have a conversation with it. I use this example all the time, but my 12-year-old is obsessed. Obsessed with anime, and she like, no, nothing else exists in her world right now other than anime. And she has created an entirely new ecosystem of anime worlds from her favorite shows using chat GPT. I mean, it’s so good that I’ve actually considered. Finding a, a publisher to have it published as fanfiction because it’s that good.

And she doesn’t type into it. She literally has a conversation with Chat. She calls it Gee. And she will say, Gee, I’m thinking about this. I want the guy to do this, and I want the girl to do this. And like she has a whole conversation and it creates this world with her that… it’s fascinating to sit and listen to how she’s using it.

So it’s the same kind of thing. Have a conversation with it. You can do it via voice, you can do it, you know, by typing whatever is easiest for you. But have a conversation with it and teach it just like you would an employee. It’s gonna learn faster. It doesn’t sleep, it doesn’t need to eat. It doesn’t need to work out.

It doesn’t need to take a break. It doesn’t, it’s not going to pause for meetings. You can have stuff running in the background while you’re doing something else. I mean, it’s the more time you spend training it, just like with a human being, the better it is.

Chip Griffin: Yep. And I’m gonna be honest, it’s gonna be, it’s gonna be more work and stress in the short term for you.

Gini Dietrich: Sure.

Chip Griffin: Yeah. I mean, this is not,

Gini Dietrich: mm-hmm.

Chip Griffin: You know, this is not a quick fix. It is not. It is not something where there’s some magic formula. You’re gonna have to try to figure out what works for you and for your team. What works for your clients. And the client piece is really where you need to start with this.

You need to spend some time thinking about what are your clients really hiring you for? What are they going to need you for 12, 18, 24 months down the road? Then start figuring out how these tools can help you to get there. Because there’s just, there is too much of this “Well, you know, I need to, I need to protect my billing model, and so I need to do value pricing because of AI.”

That is not the answer. Although if it were, what you would discover is that, that people are valuing less what you are doing today. So if you’re truly going to follow value pricing, that doesn’t mean that you get more. It means you probably get less for a lot of these things because they realize, you know, that drafting of a press release, I actually can get that out of Claude pretty well.

Gini Dietrich: Yep. Yeah.

Chip Griffin: Particularly if you feed it in your last three or four years worth of press releases.

Gini Dietrich: Yep.

Chip Griffin: It’ll be pretty darn good at coming up with them on their own.

Gini Dietrich: Yep.

Chip Griffin: Probably candidly, in less time that it takes to communicate to your team that they want it.

Gini Dietrich: Yep. 100%. Yes.

Chip Griffin: Now there will still be companies that are happy to outsource it generally.

Gini Dietrich: Sure.

Chip Griffin: Right. That’s, that is always going to exist. But the way that they value it from a price standpoint and the other things that they want alongside of it will absolutely change. And you need to be thinking about that. Because AI is not just changing how your business operates, it’s changing how other businesses operate.

It’s changing how the media operates. And so it is truly a disruptive force that we need to be thinking about as communicators and as agency folks because it, it upends a lot of what we have done, tactically at least, in recent years and over the decades. It does not upset the outcomes that are being sought after.

Gini Dietrich: That’s right.

Chip Griffin: From the work that we’re doing.

Gini Dietrich: That’s right.

Chip Griffin: And, and we lose sight of that for the tactics too often.

Gini Dietrich: One of the things that I did is I built an agent, and I call it my co CEO. And as I was building it, I was going through a really rough HR time and so I used it mostly, honestly, just to vent.

But it got to know me, and what’s important to me, and my voice, and what things I wanted to be human forward on, and what things I needed to stay professional on. And so I, as I was building it, it was, I was going through that process. Now I can say to it, okay, we’re thinking about doing this. So for instance, a client came to me probably midyear last year and said, Hey, we really want your team to do an audit of all of our brands and where they sit on the PESO model maturity level.

And I kind of laughed and said, well, I can tell you right now, they’re all at level zero. And he was like, great, that’s good to know. What’s what takes us from zero to one, one to two, and so on up. And I thought about that for a little bit and I was like, hmm. I don’t have an answer for that. And so I went into my CO CEO and I had a conversation with it.

Like, if we were gonna build a maturity model for the PESO model for an enterprise customer, what does that look like? And it probably took two weeks for me to get something that I could go back to him with and feel comfortable and confident with it. But it would’ve taken me two months to do that on my own.

So, you know, it helps you think, it helps poke through holes in things. You have an AI that you’re building and I hope it’s okay for me to mention this ’cause I don’t know if it’s available yet, but I got to beta test it and it’s, I put in there that I was looking for. I said, okay, this is where, this is where the business is at the end of 2025.

These are our goals for 2026. Here’s what I’d like to do in the next three to five years. Here’s like, I put in all of that information, where are the holes? And it started poking holes into things that I had never even considered. And I was like. Chip, this is really good. It’s just, it’s really, really good.

So when you, when you train it, when you teach it what you’re wanting, what your voice is, what you’re trying to achieve, it is going to help you in more ways than one. It’s gonna help you think through problems. It’s gonna help you come up with solutions you didn’t consider. And like I said, it doesn’t need to sleep.

So it can work in the background while you’re doing other things.

Chip Griffin: Yeah. And there are a lot of these ways that we can innovate for our businesses and that particular example, it is live on the SAGA website now. It’s an AI agent called Sage.

Gini Dietrich: It’s awesome and everyone should check it out.

Chip Griffin: It’s trained on a huge volume of my both public and private materials that I’ve created over the last eight or nine years, and it does a remarkably good job of mimicking the advice I would give. Is it a hundred percent? One-to-one? No. Yeah, but it’s, it’s pretty darn close to the point where I’ve had a couple of clients now who have tried it and then asked me the same question they asked of Sage, and they got almost exactly the same answer.

And, and so that’s how, you know, it’s, it’s working pretty well because I think, as any listener knows, I have some views that are not necessarily exactly in line with every other advisor in the agency space. And so, and in some of those cases, they were pieces of advice that you wouldn’t get if you went somewhere else.

So, you know, you can tell that it’s actually using the training materials. And not simply doing a general knowledge search. But these are all things, it does take time. You’ve gotta have the material to provide to it. You need to spend the time with it, as you did in conversing and going back and forth.

But the more you go back and forth, the smarter it gets.

Gini Dietrich: That’s right.

Chip Griffin: And the better it can help you the next time something comes along.

Gini Dietrich: That’s right.

Chip Griffin: And I think the other thing is that the more you use these tools, the more it forces you to think about some of these things. Because in order to get the most from them, you really have to be very clear about who is your ideal client?

What are the services you provide? What is the value you deliver? And so, it’s just like a business plan. I always say that the business plan itself doesn’t really matter, but the process you go through to create it does.

The process you go through to train your AI itself is beneficial and helps to get clarity. Because the clearer you are with the AI, the clearer you are with yourself by necessity. And so you need to be thinking about these things.

You need to be really thinking about making much more radical change to your business over the next year or two than you probably have previously thought. You really need to be thinking about how not just technology, but client needs will force this change, otherwise you are gonna get left behind.

Gini Dietrich: Yeah, I totally agree. And to your point earlier, if you evolve and if you use it, and you’re better, you’re doing a better job of understanding what it is that your clients are willing to pay for, and they’re still willing to do it. They just don’t wanna pay for social posts and news releases.

Chip Griffin: That’s right. I mean, there’s a huge opportunity here. There’s a giant threat,

Gini Dietrich: huge opportunity,

Chip Griffin: and I don’t wanna minimize that, but there’s a huge opportunity.

Gini Dietrich: Yep.

Chip Griffin: But the key is you actually have to evolve and change. You can’t just play buzzword bingo.

Gini Dietrich: Yes, please.

Chip Griffin: Just slapping AI on top of something that you deliver that’s not gonna help you.

Gini Dietrich: And it’s fun. It’s fun to test it. It’s fun to try it out. So do it.

Chip Griffin: Yeah,

Gini Dietrich: Do it, do it, do it.

Chip Griffin: I mean, but we can’t minimize. It is scary for a lot of people too. I mean,

Gini Dietrich: sure, absolutely. Yeah.

Chip Griffin: But you’ve gotta, you’ve gotta embrace that fear if you wanna succeed.

Gini Dietrich: Yeah. I always say when I have somebody new join the team that’s scared of it, I say, all right, let’s do it.

Let’s use it for something personal. So I will say that to you as well. Meal planning, fitness, hobbies. Managing your kids’ meltdowns, whatever it happens to be, just try it for something. Write a poem in a birthday card. Try it for something personal, and I guarantee you, you’ll be hooked.

Chip Griffin: I had no idea we’d be getting to poems and birthday cards here today.

So I think that’s the note that we’re gonna wrap up this episode on. I’m Chip Griffin.

Gini Dietrich: And I’m Gini Dietrich.

Chip Griffin: And it depends.