AI Ethics in Marketing: Why Strategy and Responsibility Must Go Hand in Hand

0:08Hello and welcome to another episode of the duct tape marketing podcast. This is John Jance. My guest today is Paul Chaney. He's a seasoned digital marketer, B2B writer, and editor. He is the publisher of the AI Marketing Ethics Digest on Substack and a longtime voice on responsible tech adoption. With a background in content strategy and digital ethics, Paul explores the psychology and structural effects of AI on the modern workplace. So, Paul, welcome to the show. Thank you. It's a pleasure to be here, John. These days, you know, I talk about AI pretty much in every episode, it feels like. But I've not had anybody on this topic, so I thought it was really an interesting one.

0:51I'm curious, is there anything that sparked your interest? I mean, you've been around as long as I have, practically. Anything that sparked your interest in AI? You know, I think if you go back to, let's say, the early blogging days, you know, which I know both you and I remember, as it began to be incorporated into the business environment, you know, there was this maybe sort of ad hoc approach. And finally, we got around to, you know, promoting the idea of having blogging policies, that kind of thing. It was just guardrails for safety, making sure people didn't say stuff that they shouldn't say and all of that kind of thing.

1:26well, you know, we've jumped through social media hoops and things like that now. And it was the same thing with that. Well, now we're on something altogether different, a whole new layer of this kind of what could end up being a real mess if organizations don't manage it effectively and cohesively. But in terms of my own interest, I will tell you this. I looked at what was being talked about as it began, you know, as AI began to, you know, become really a central part of conversations. I saw marketers talking about the tools, shiny new tools. I saw AI ethics people talking about ethics. What I didn't see was an integration of the two, a synthesis of the two.

2:13I didn't see, and I felt like, well, here's a hole that needs to be filled or a gap that needs to be filled. And I don't see anybody else doing it. So what the heck? I'll do it too. And because I'm not really a video kind of guy, I decided to do a newsletter. And that's when the, in August of 2023, the AI Marketing Ethics Digest was born. And I will be honest with you, completely candid. It was not a topic that I was even all that interested in. Certainly not passionate about at the time, but I just felt like, yeah, we need to be talking about responsible AI. And because my field is marketing, I said, well, let's just sequester it to that for now.

2:57Well, a lot of organizations, I'm sure, are wrestling with this or will be wrestling with this from a compliance, HR, legal standpoint. But I also think we look at things with that lens. You know, what's the impact on the company or why would they do this? Because they're forced to do this, right? But I think sometimes we forget about the actual consumer that is out there, you know, the buyer, the, you know, in B2B situations, you know, the business that's out there. How do you feel like, how are we going to be responsible to that thought, you know, and not just a compliance thought? Well, and yeah, compliance, you know, you think about legal and all of that and then, you know, regulatory policies and stuff.

3:41But I think first and foremost, as with anything, you have to think about your customer or the consumer. What's fair to them? What would be honorable? What would be moral? What would be ethical where they're concerned? And I think you have to approach this whole use of AI from that standpoint. Now, does that mean that you disclose every use of AI that you're involved with and everything? I don't necessarily think so. I do think at the very heart of this, though, lies the ethic of transparency. You know, that's implicit across the board. And, you know, the big talk that you see quite a lot is about bias.

4:23These LLMs are trained on, you know, certain information and they're going to spit that out, right, as they've been trained on it. And I think those kinds of things can affect your relationship with a customer. And it can also do damage to your own reputation as a brand or a business if you're not careful, if you're not making sure you're auditing, you know, what's being put out there, you're monitoring it, you know, that kind of thing. So I think first and foremost, our responsibility is to the consumer and making sure we're treating them right where this is concerned. Well, you know, I've been saying for years, probably 30 years, I started saying, you know, strategy before tactics.

5:06And, you know, I've kind of been changing that tongue in cheek, you know, strategy before technology, you know, because everybody's just jumping into these tools and saying, look, this can make me do this faster or this can make me do this more efficiently, you know, only to come to find out that they're doing like really bad work faster or at scale. And so that's, you know, to me, you know, I always analyze every new thing that's come along. I mean, the web came along and social media and all these mobile devices and, you know, now AI and just look at it. Fundamentally, what are we here to do as marketers?

5:39You know, let's develop a strategy that does that. And can the tools allow us to do that? You know, as opposed to, I mean, nothing drives me crazier than seeing people show these whiz bang things. Like, you know, somebody invited me to be on a podcast and they were going to have an AI bot interview me for the podcast. Oh, my gosh. It's like, OK, it can do that, but should it? Do you really want to do that? Just because you can doesn't mean you're good. Yeah. And I think that's what people do. They get really enamored with like, look at this cool thing. So people are now out there building AI agents, you know, for everything without any thought about like, what's the strategic impact of this for my business?

6:18And I think ethics runs really side by side with strategy, quite frankly. Yeah. And I will have to say, I mean, I've been doing this newsletter now for a year and 10 months, and I just crossed the 700 subscriber line. You know, you think, well, gee, a year and 10 months, you should be further along than that. But that's kind of telling me that maybe this is not top of mind with marketers right now, the ethical side of things. I mean, how many newsletters and blogs and YouTube videos are out there about all of these tools and stuff? That's what folks are focused on. But to your point, if you don't take this from somewhat of a top-down or strategic mentality, well, one of the things you're going to end up with is a lot of shadow AI.

7:03You're going to end up with people using this, not disclosing they're using it or full disclosure. You may have this department using this, this department using this, and there's no governance whatsoever. You run into privacy risk. You run into safety risk, I think, with all of this, too. We think about, you know, cyber tech and all of that stuff. So I think there's a lot of ways that this can go wrong if you're not keeping ethical guardrails in place. And it's not like it's rocket science. I mean, a lot of it's just common sense, you know. And so establishing things like an AI ethics council or committee to oversee a lot of this, and then maybe some kind of usage policy that we can do this, we can't do this.

7:51Not unlike we did with blogging and social media. Yeah, I'm starting to see some parallels with a lot of organizations we work with when people started posting on social media with their personal phones. Right. So, I mean, it's kind of the same thing. It's like half the people in the organization have a personal ChatGPT account, you know, but how is it being used? How is it being monitored? How is it being retained? Right. Because, you know, when they leave the organization, all of that, whatever the work they did on behalf of the organization leaves with it. So I think it's really just a set of kind of common sense policies in some cases or processes.

8:29Yeah. In a couple of weeks, I'm going to be publishing an article on the newsletter. I call it the the Tower of Babel problem in corporate AI. And, you know, everybody's probably familiar with the Old Testament Tower of Babel, where they were building the tower and then everybody started speaking a different language and everything stopped. Right. You know, well, I think that can happen in the modern workplace, too. And I just like I mentioned, shadow AI everywhere, maybe duplicate spending, inconsistency in the voice. Marketing says it one way. Some other department says it another way, you know, all of these kind of things.

9:05And what's that going to do? That's going to impede productivity. And I think, you know, that, too, has ethical sort of parameters around it. But if you start it with a strategic mindset, understanding that it's, you know, you're not going to roll it out in a full flesh sort of way across the enterprise. You know, you're going to start small. You're going to start with pilot projects, maybe in a really high impact use case. Make sure that's working like it should. Then you gradually expand it. But it's all done under some kind of governance framework where there's approval, there's risk checks, there's all of that kind of checks and balances, if you would, right?

9:45Is anybody producing, you know, that kind of policy? I mean, you know, you've got people out there that will come in and audit your security and people that will audit your HR practices. Is anybody doing that now as a service? I'm sure there are. And I wish I had in place some real bona fide kind of, you know, examples or case studies. That's one thing I am working on, by the way, but we're still very early in this. Sure, sure. Even though it's been out since, what, 2022, you know? Yeah. I mean, it's been out, AI has been out much longer, but the practical use of it in our case is different. So I don't have anything on hand, but just, you know, I'm not here to plug my newsletter, but I would say just keep reading the newsletter and you're going to see case studies.

10:28because I'm keeping my ear to the ground on all that stuff. One of the things I think you posted on LinkedIn that I'm sure came from the newsletter as well, or maybe that's where I read it, but the idea of this generative AI pyramid. You want to unpack that concept? Yeah, a funny story about that. I interviewed Charlene Lee recently for the newsletter, and she had used this Maslow's hierarchy concept and applied it to generative AI, and her focus is leadership. And I got to thinking, you know, let me see if I could do something like that where it comes to the ethical side of things. So I did. I went to Chad GPT or maybe Claude.

11:08It's Chad GPT, Claude and perplexity. That's my trifecta. And I said, OK, can we create something that is, you know, using that model? And but it has this ethical orientation to it. And sure enough, it did. But I call it the generative AI business adoption hierarchy. I'm not one for flamboyant names, John. I'm just a Plungent guy. But it starts basically, it is kind of follows Maslow's in a sense. It starts with the lower level awareness and access. Does my team know what these tools can do? Are they allowed to use them? Have they been trained on them? And then the next level, if you want to think of it in a hierarchical fashion, is security and trust.

11:52Do we have guidelines in place to make sure we're using this safely and ethically? And I have a friend who's very well invested in AI technology who says, no, Paul, you got that backwards. You really need to start with those clear guidelines, make sure that governance is in place. So, you know, you could perhaps switch those two around. But once you've got that, then that's really where the magic starts. That's where you can begin to see productivity take place because you're using it in a way where there is the governance, where there is approval, where there's tried and true tested sort of things through these various pilots, that kind of thing.

12:32Marketing is here's your prompt stack, if you want to use that term. HR, here's yours. Finance, here's yours, et cetera. And as you get to that level, then you're going to begin to leap over into the level where it really does move from tactics to strategy. And it becomes a part of the business model itself. And I don't know if it's sort of an epiphanal moment or if it's just a gradual change of the sense of culture where AI begins to be baked into the organization. And that's kind of that like Maslow's self-actualization. That's what that is. Now, how long does that take? I don't know. I mean, it depends on the organization and how flexible they are and open to transformation they are.

13:19Well, I have a theory. Yes, sir. I think that somebody is going to create the AI system in a box. So they're going to walk into an organization like they would an ERP software or something and say, here it is. It's built and branded just for you. It's got your policies in it. you know, it's$800 a month or something like that, you don't have to figure it out now. Because what I'm seeing a lot of is, you know, everybody's going, oh, this new tool, that new tool, I should try that, you know. And I think businesses are already exhausted, you know, over the trying to figure it out. So then there's obviously AI agencies and consultants, you know, that are cropping up that are actually kind of creating the agents and things for people.

14:01But I think there's going to be a turnkey solution or fairly turnkey solution that kind of handles all of that stuff that comes out. I don't know, maybe it'll be a startup. Maybe it'll be the big three. Yeah. Hey, when do you want to get, let's get started on our... Well, that's weird. We're actually headed that direction with a tool for marketing. So we're going to walk in and license actually a fully built, a fully branded system for an organization and then teach your people how to use that one system as opposed to everybody just kind of doing their own thing and building their own custom GPTs.

14:34I think that's the only, that's one of the ways to maybe, well, yeah, I make it accurate, but put some guardrails on it. I'm in, I'm done or well, it's in place. I'll put it that way. I haven't had any takers as of yet that, that sort of thing with the ethics side of thing. And I've really, again, kind of focused on marketing, but I think what I've created could, you know, branch out into other departments and be used in other departments. And so I think you're absolutely right. There will be a package solution for this. And I think, you know, again, the big companies are just going to hire McKinsey and say, you know, do this for me, you know, but it's that, you know,$10 million electrical contractor that's like, well, I know I need AI, but where do I even start?

15:18You know, and so I think there's a right market for, you know, somebody to kind of hit that kind of mid market for. Yeah, I agree. So there's another term that I grabbed from some of your writing, AI techno stress. And we were just talking about it. I see a lot of stress from business owners right now. But how does that come into play? I mean, are people, when somebody is full-blown techno stress, are they just shutting down? Or what's going on with that and how you're applying that to the ethics component? And as I was, you know, as the newsletter has grown and evolved and it has, one of the things that sort of began to, I began to think about and, you know, give some, give some thought to was this idea of the stress element that is involved in learning to use these tools.

16:11And maybe being forced to use these tools, even when, you know, if say you've been a long time employee and you've just done it a certain way for years and now you got to do a different way. Well, that's going to be a stressor. But look, techno stress. And the term was first coined in 1984, I believe, long before AI came on the scene. You know, you forget a password and you're having to go redo a password. That's techno stress. You know, the Internet goes down or whatever the case may be. But AI is just like a whole other layer of this. And I think it's inducing a lot of stress at a lot of levels.

16:47You mentioned the business owner, the$10 million electrical company. Surely that business owner has, you know, some confusion related stress where that's concerned. But maybe you're a frontline worker or a knowledge worker and you're sitting in front of your computer every day. And now you've been introduced with this new set of tools that you got to learn how to use and get a grasp of. And I can't tell you how many times I've seen people, maybe on LinkedIn commenting, other places saying, well, you know, I tried chat GPT, but it didn't really do anything for me, you know, and I'm thinking, well, you're, you're lacking fundamental training that needs to be involved.

17:23And so that certainly is a part of the solution. But I think to me, and I've really gone down that rabbit trail now with the AI techno stress concept, I'm actually going to be publishing a book here later this year, focusing on that. And, you know, I just feel like that's an element that ties to the ethical use of all this. There is ethics involved when it could cause some kind of emotional or mental or psychological harm to your employees. That's ethical. So that fits, you know, it's still all under the same umbrella. Well, it's interesting because, you know, I know a lot of employees get surveyed and, you know, the probably above like I want more money is like I want to know what it takes to do my job.

18:12I want to have the tools to do my job. And I hadn't really thought about that, but a lot of organizations are just saying, hey, figure this stuff out. And that, you know, lack of training is actually a real detriment to the organization, isn't it? Right. And I think, you know, to your point about the box to solution will help decrease that stress level. But let me give you a quick example. I was in a meeting yesterday with an agency. My day job is writing and editing for B2B. And I work with an agency out on the West Coast. And they have been around 30 years. Their founder, she's probably in her 70s.

18:51And they've done great work. but they'd done it their way and now they're being confronted with having to do it a different way so much so that the founder has decided she wants to go ahead and retire she's going to turn it over to her second in command so to speak and uh that lady and i were talking recently about this whole ai thing and you could see the curiosity in her mind but you could also see you know some bit of conflict because they have very strong detailed policies and ways of operating. And now, you know, the way they want their writers to work, for example, and now that's all being challenged.

19:31And the conversation in the meeting yesterday was with all the writers. I'm again, one of them, and it was AI this and AI that and AI the other. And, you know, so I think, again, it goes back to the organization, how, how flexible are you willing to be? How progressive are you willing to be just in every case? As in every case, there's going to be a bell curve. There's going to be the innovators, early adopters. There's going to be, you know, the early majority. There's also going to be the laggards who just going to say, nope, we're not doing it. And those are the folks that I'm very concerned about are going to lose in the long run.

20:07Well, no question. I mean, there's absolutely no question. This is not a time to bury your head in the sand, but You know, it's interesting. I hadn't really thought about that, but I bet you there will be a wave of people, business owners, they're just like, forget it. I just don't have the bandwidth or the energy to figure this new thing out because it so fundamentally shifts how we go to work. You know, what an org chart actually is going to look like. I mean, so it's not like, oh, I got to figure out Twitter. It is really, I've got to like replumb my entire business here. I think that probably will actually wipe it, not, not wipe people out in the sense of, of, you know, being overtaken, but literally just saying, I, I, you know, enough, I give.

20:47Well, you know, and you're so smart in taking a leadership role in this because, you know, you, you've been around a long time. You've, you've got a lot of influence and I, people listen to you and, and you, I know, I don't know if, if your business has changed over the years, but, you know, you focused on small business, or at least you used to, I don't know if it's larger now, but. You know, I think about small businesses. I come from small business. I am a small business person and my heart still lies there. And I can imagine a lot of these are just a deer in the headlights look when it comes to this stuff.

21:20And they have no idea about where to go, what steps to take. And so what does that mean for somebody in a consultant kind of role? Hopefully somebody could come in and say, all right, we can figure this out. Don't worry about it. Here's a plan. And like you say, kind of a box solution. Yeah. Yeah. Well, you asked if my business has changed. I kiddingly say we're in perpetual beta. Yeah, really. I understand. So, Paul, I appreciate you taking a few moments to stop by the Ductate Marketing Podcast. Where would you invite people to connect with you, but probably also if this topic interests you to be able to subscribe to your newsletter?

21:59Yes, sir. It's AIMarketingEthics.com. That will redirect you to the sub stack. My day-to-day business is prescriptiverriting.com. I don't write prescriptions, but anyway, it's B2B writing. And you can always find me on LinkedIn. So that's where I hang my hat a lot. And that and Substack these days, I'm not doing much with any of the other social networks anymore. Yeah. Awesome. Well, again, I appreciate you stopping by. Hopefully we'll run into you one of these days out there on the road. I would hope so, John. And thank you again for this opportunity. I appreciate it. .