Is All AI Content Slop? What Leaders Need To Know
AI content becomes a trust problem when it simulates experience or authority a brand has not earned. Considerateness offers a better standard.
Get the next practical AI marketing episode wherever you listen.
AI content is not automatically slop. It is also not automatically good because it saved time. The right question is whether the work is considerate of the person receiving it.
That sounds simple, but it gives marketing leaders a more useful way to make decisions than either extreme. A blanket ban on AI can throw away a helpful tool. An all-in approach can quietly erode trust, especially when a brand is asking people to believe it, learn from it, or buy from it.
Slop is a trust problem
We use the word slop for cheaply made content with little thought behind it: bad images, empty writing, videos that sound polished but say nothing, and synthetic material presented as if it came from a real person. The problem is not merely that the audience can spot it. The problem is what they conclude when they do.
In a conversation with a friend, we discussed an apparently impressive AI-generated reel that fell apart under closer attention. The script was generic, the presentation appeared fake, and the creator's credibility fell with it. Views did not solve the trust problem.
The harmful end of the spectrum is clear. Fake doctor endorsements, made-up testimonials, invented people, and fabricated authority are deceptive. They use the appearance of human proof to get a result that the brand has not earned.
Most teams will face subtler choices. A Father's Day post built around AI-generated people prompted a useful internal discussion at my company. The right response was not to stop using AI everywhere. It was to clarify the boundary: for a high-trust brand, people in marketing should be real people. An AI-created object or illustrative element may be fine. A synthetic person changes the relationship with the audience.
Context determines the standard
Entertainment has a different expectation from education, product research, health guidance, or a personal brand. A story about a tire chasing a Jeep can work as AI-assisted entertainment because the audience is there for the story. The same tolerance does not automatically apply to a review of a backpack, an expert recommendation, or advice with real consequences.
When I research a product, I want to know whether the person talking has used it. A faceless summary built from everyone else's opinions cannot offer that. The content may look efficient, but it does not give me the human experience I came for.
On the other hand, AI can be perfectly useful when the job is curation, structure, or visualization. An AI-curated news feed can be valuable if it works, helps people find what matters, and does not pretend to be something it is not. Educational visuals can also make information easier to understand. The use should fit the reader's expectation and the stakes of the decision.
Ask what is most considerate
Considerateness is a better compass than trying to identify every technical boundary. Before publishing, ask: What does the audience expect here? Does this use of AI preserve or weaken that expectation? Would a reasonable person feel misled if they knew how it was made?
That question should shape brand rules. A high-trust business may decide that people are always photographed, that testimonials are always verifiable, and that sensitive advice has a clearly qualified human source. A team producing entertainment may set different rules. Both can use AI well if their standards are visible and applied consistently.
Leaders also need to avoid becoming the approval bottleneck for every social post. Give the team clear boundaries, let people work within them, and use misses as an opportunity to make the boundary more precise. One imperfect post does not require panic. It can reveal where the policy needs more clarity.
Use AI to improve the work, not impersonate the source
AI is excellent at assisting with a draft, an idea, a visual direction, a research summary, or a tedious production step. It gets dangerous when the output is used to simulate the human experience, expertise, or consent that gives a message its authority.
Keep the human value in the work. Let AI increase the range of what your team can create, but require the message, judgment, claims, and trust signals to hold up when someone looks closely. The content that lasts will not be the content that merely looks human. It will be the content that treats humans with enough care to earn belief.
The conversation delves into the overwhelming nature of AI news and the impact of AI slop on marketing leaders. It explores ethical concerns and the balance between AI content creation and trust. The discussion also covers the perception of AI content, its impact on credibility, and the consideration and pragmatism in AI content creation. Additionally, it addresses the most considerate approach to AI content and ethical considerations in AI-generated images. The conversation delves into the complexities of navigating credibility and trust in AI, highlighting considerations, ethical dilemmas, and the impact of AI on medical advice. It then explores the shift in perception of AI-generated content, emphasizing authenticity, vulnerability, and the evolving role of AI in marketing. The discussion concludes with a focus on pragmatism and consideration in AI usage, emphasizing cost-effectiveness, human value, and the ongoing evolution of AI usage. The hosts also encourage feedback and engagement with the audience, inviting them to reach out on LinkedIn.
Takeaways
- Ethical concerns arise from the use of AI in content creation.
- The most considerate approach to AI content creation involves balancing practicality and trust. Consideration and ethical evaluation are crucial in AI usage
- Recognition of AI-generated content is impacting content creation
- Pragmatic approach and human value are key in AI usage
Chapters
- 00:00 The Hiatus and Overwhelming AI News
- 05:53 Impact of AI on Design and Content Creation
- 11:47 Impact of AI Content on Trust and Credibility
- 19:34 Ethical Considerations in AI-Generated Images
- 28:19 The Shift in Perception of AI Content
- 35:28 Pragmatism and Consideration in AI Usage
- 41:13 Closing and Call to Action
Dan Sanchez: Welcome back to the AI Driven Marketer. I'm Dan Sanchez. Friends call me Danchez, and of course I'm joined by Ken Frere this Friday.
Ken Freire: What's up everyone? What's up, Dan?
Dan Sanchez: And man, it is good to be back. Like I said last week, I was kind of on a hiatus and it wasn't like a planned one, otherwise I probably would have announced it to y'all and be like, Hey, I'm gonna take the summer off. I think I just stopped uploading for a while. I think we all go through that, you know, you're kinda like grinding, grinding, and then you're just kinda like, I'm out for a bit. I've done it with LinkedIn, I've done it with this is probably the first time I've done it with this show, but you just kinda get to the point where you're like, I don't I don't even know what to say this week. All this AI news, it's just chaos. And I I I I don't have an opinion. I'm also just grinding working on other things now.
Ken Freire: Well, it's all overwhelming too. Yeah, but it's all overwhelming 'cause there's just so much is happening so quickly. You're like, What do I work on? What do I do? How do I make it happen? And you kinda just need to focus on the main priorities.
Dan Sanchez: So did you even did you listen to the solo I put out last week? Awesome. That helps. I'm like, wait, I might have to catch you up, bro. so again, pivoting the show more towards marketing leaders, but I think this will be applicable to a lot of people. today we're gonna be talking about a conversation that's just come up recently and hit hard in my I the company I'm working for. Like this was a big conversation. And I'm like, dang, we should talk, I should talk about this on the pod, because
Ken Freire: I did. I'm a fan, bro. I'm your number one fan.
Dan Sanchez: I think marketing leaders are probably facing this issue and a lot of people talk about parts of it on LinkedIn posts, but I don't think they're they're either I don't know, they're like they're hitting it in too precise a way where I feel like there's a lot of nuance to this particular topic. And so we're gonna kind of dive into it and talk about the degrees when it comes comes to AI content and specifically AI slop, right? That's the derogatory term everybody's throwing around on essentially AI content that's just I don't know, like eat cheaply made, no thought to it, just thrown out there, or is just like obvious, not great AI content, whether it's images, videos, written word, whatever it is. If it's made by AI, people throw that term out when it's bad AI content of any kind.
Ken Freire: Yeah, and I I mean, I can tell you from my vantage point, especially when it comes to images, I tend to create AI slop images. So I have to be mindful of it because I'm not an expert at graphic design. So everything always looks crappy. So I try not to do those.
Dan Sanchez: So today I wanted to talk about it because it like it actually blew up. Like we actually had like a emergency meeting. I'm gonna be a little vulnerable with some of the things that happened internally, but we did work it through. We had resolution and we're on a path now. But it's because of this internal conversation that I'm like, I feel like I had to really help like lead the charge on like where to st where AI starts and where it stops internally in order to make the most of it. Because I think this is where we have to actually wrestle as as marketers and marketing marketing leaders, because it's not all in. You can go all in on AI and use AI for everything and have your agents spitting out all this stuff. and did you hear like the world's like first solo person billion dollar company? that like that just happened. Like they talked about it happening for years and it finally it happened. There was a guy who's doing some kind of like What is it, pharmaceutical weight loss? It's the you know, the miracle weight loss drug, the GLP ones. Yeah, he's doing stuff like that, but spinning up a whole website, but he's using AI for all of it. And that's how he got he got to a billion in in value, right? Over a billion. And the New York Times did a big story on it. What the New York Times actually failed to report on that Russell Russell Russell Brunson hit, which is out of all people to call him out for this. I'm like, Russell, you're not exactly clean, my man.
Ken Freire: Yeah, the GLP. Yeah.
Dan Sanchez: Anyway, Russell called him out for the fact that almost all his stuff was AI slop, including his doctor endorsements, including all the testimonials were just completely false and fake. So that's on an extreme end. It's like, yeah, he's making a money a lot of money, but he's doing it unethically. Completely unethically. I mean, when you're faking doctor testimonials and just coming up with headshots of doctors and making up the names and they'd find all these hundreds of doctors and some of them would be like Professor Dumbledore, you know, the kind of doctors. That's how they knew they're like, these aren't real people. They'd have Facebook profiles, but they'd be like fake Facebook profiles and stuff. So, and then a bunch of the testimonials, almost all the testimonials were completely fake. So there's a reason why people are blowing back on this kind of stuff because, you know, it doesn't have a good look. When people dig into it and find out you're kind of a fraud, you know, the AI thing can get out of control. But At the same time, like you said, it's a really useful tool to be able to come up with a little graphic of some kind, a little image flyer without having to go to a designer. Shoot, I am a graphic designer, and I find that ChatGPT is often doing better than what I could do with three hours of work. And I mean to come up with a flyer with lots of text on it and it still looks good. AI can do it better, faster than if I had opened up Photoshop and just done spent three hours on the task. As someone who knows how to organize things visually as somebody who knows how to use Photoshop well. It's better than me now. I'm not an A plus designer. An A plus designer will still crush ChatGPT. It still takes them longer. but even for people who are pretty good at design, I say I consider myself a B level designer. It's still better. But it still looks like AI. So then how do we navigate these quandaries? And that's the big conversation. So we're gonna talk a lot about slop in this episode, but I want to tell the story. Essentially I had a post I I didn't even see it. It's something my my marketing manager posted maybe two weekends ago. It was a Father's Day promotional post. And it was picture of like a daughter and a dad sitting on the back of a pickup truck with like a tagline in it. And it got posted on Saturday. It a bunch of people in the company were like, What the heck is slop kinda is this? And it turned into a big long meeting the next the next week. And I like, gosh, all right, here we go. And we had to talk about how far we would go. with AI. At first the the initial reaction was we have to stop using AI ever in anything. And we need strict approval processes. They like overreacted and everything. I generally take the position, I have a rare position as a marketing leader where most marketing leaders are not comfortable with this, but I I feel like I'm still right. I give my social media team and my content team the ability to write stuff and publish things. I give them boundaries of what to do. And as long as they stay within those boundaries, I let them post things and I don't have to review it. If they're wondering, they can check in with me, but generally I let them post. I don't I don't and then I if they post something or I I think it was like could have been better or whatever, then I tell them after the fact. I'm like, hey, this post could have been better. You should have done this. Next time do this. Because I find if you have a approval or process, you become the bottleneck and you don't get stuff out fast enough. They can't react fast enough. You know, 'cause you're marketing leader, you're in meetings and stuff and you can't always like get to the things and I got my task list I'm trying to work on. So I just let them go. so this one went out and I pr if it had been an approval process, I wouldn't have approved it. Let's put it that way. But I gave her the creative ability to do it. She did it. And I'm really thankful that she did because sh most of the time she does really great. But this one was a miss and it was okay. But the overreaction to it was like, my gosh, we should stop using AI altogether.
Ken Freire: Yeah. Well, but and Well, I mean, I going back to what you were just saying when it comes to giving people the approval, what I have found, I I have a marketing coach and he says something really profound to me the other day. He's like, What, maybe three percent, five percent of your audience will see it, and you're gonna be this whole bottleneck. So he's like, just keep posting. Within two weeks, everything's gonna blow over. Most of the drama is always gonna be internally, thinking that it's a way bigger deal. Most of your people are not. And and this is I'm talking about general posts. Now, if you put like a really conscious Controversial hot take that might be a little different, but I'm guessing this wasn't that much of a controversial.
Dan Sanchez: Yeah. Yeah. No, it was just one where it was obviously AI. I had already made a policy, like a boundary, of we don't do AI people. Not at this brand. It's a high trust brand. And there's something about AI people that just to me that was like an a guideline that I put in place that I'm like, they had had some AI people on the website and I'm like, this just it's not a good look. So anytime we use people, it's it's gotta be real people. photo like photoshopped in cut out and put into the thing. So if you go to our YouTube channel, we have lots of people, but they're always photoshopped in and the the object image that emphasizes the what the show's about will might be AI, but the people are always real. So this was an AI and she kind of justified it because it was only the back of their heads, you know. But I was like, like I see where you were thinking there. That's cool. Just don't do use people. Like don't use people at all. That's let me let me clarify the boundaries. That's all that's all that really needed to happen. I just needed to reclarify the boundary. Don't do that. but of course the overreaction happened. And so it became a really good conversation though, around where do we use AI and even like we actually discovered that there were some places where I had been using AI and they thought it was great. So it becomes a question of like, all right, so where do we use AI and where do we not use AI? And I think that's the thing that marketing leaders are probably wrestling with the most. Because they're seeing all this stuff out there about like, you need a like one person can do the work of ten people? It's kinda like, yes, but like there still has to be nuances in how we're navigating this new tool called AI in order to do our content. So is AI content always bad? No. Is it always good? No. So there's like middle ground and I find that that's the conversation we had internally in order to figure it out. so I have to ask you the question. How do you how do you have you using Content on places like LinkedIn or Facebook or ads that you're like, that was good. And have you seen content that it was AI and it was bad? Have you seen both out there?
Ken Freire: yes, I have seen really good AI content that I was like, wow. And then there's others that I've seen that I was like, this is really bad. It I and I'll add another one that I think sometimes is is more tricky is and let me give context to this. I think that within our spheres, right, like marketing leaders, people who are using AI, you can spot the AI slap a lot quicker than the general population. And a great a great example of this is like just the other day, remember I sent you that reel and I was like, bro, I think this is AI slop. Like this is all AI generated. I could tell from the audio, like the script was completely AI. And then I was trying to decide, I'm like, is the video AI? And I had to like watch it several times to be like, I think this is fake. It's like
Dan Sanchez: Yeah. Yeah. I think it was. It was really good. It was the best AI video I'd ever seen. Of like podcast of like some guy on a podcast with a mic on a podcast studio somewhere.
Ken Freire: Yeah. So Yeah. And he's got a mil he's got a million followers on Instagram. That thing was hitting hundreds of thousands of views, right? Like so all of a sudden, when you say, Have you seen both? I was Man, sometimes I've seen it all in one reel, where like the video is like on par. You're watching it quick, you're not gonna catch it. But you listen to the script and I'm like, my gosh, this is this thing is completely AI generated, AI slop all day, every day. So that that's the stuff that I'm seeing now.
Dan Sanchez: How did it make you feel about that video?
Ken Freire: I lost complete trust with that individual. And especially because that person was saying he's a person of faith. And I'm like, Now now you're acting like this person who's popp popping up a lot of this great content. And you know, part of AI Slop too is like, it sounds smart, but you're really not saying anything. So all of that was coming out. I'm like, I discredited him. And the person who showed me it showed me this individual. No, like I wanna be like this AI Instagram follower or this this Instagram follower and I told them, Don't be like
Dan Sanchez: Did he say AI or did he say this Instagram creator? Okay.
Ken Freire: This Instagram creator, sorry. He wants to be like this Instagram creator. And I was like, Don't follow this guy. This guy's just using AI for everything. And it like sure he's getting views, but like that's not who you are. And he's like, man, you're right. I don't want to be that guy. I'm like, okay, cool. Like be you
Dan Sanchez: Yeah. Yeah. So there's have you seen anything good out there? Where you're like, this is clearly AI, but I am I am here for it. It is good.
Ken Freire: Yeah, I mean I've seen some some writing that I've really enjoyed. that I'm like and they they they said it like up front. They're like, this is AI, you know, and like but just the way it was written, I'm like, okay, cool. I can I can dig this. I like some AI content that I get, obviously, that is more factual stuff, like factual information. Right now I'm digging a lot of YouTube videos that are using AI like on giving history stuff and they're using AI for it. we have a friend of ours who who's who's doing a vi a channel like that, which I'm like, I appreciated that. Like the there's they're giving visuals to historical information that we normally wouldn't have. So when it comes to the visual side of things.
Dan Sanchez: Yep, yep. Yeah. I find there are times where I I see AI and immediately discredit it because it's AI. And then there are times where I'm like, I'm here for it. Kind of like especially in entertainment. If it's entertainment, don't care at all if it's AI or real. It's entertainment. Like it's the point of it is not about credibility. It's not about authority. It's just it's here to make me laugh or smile or get me intrigued or something. Like I I just don't care. So I feel like education is pretty free game. Obviously, is it different if it's real? Yeah. And it might be more intriguing in some ways. But can AI can do different things with entertainment. So it's not like AI will dominate entertainment. It probably will in a lot of ways. But it's not like the real stuff goes away necessarily. It becomes just a different. but when it comes to education, like for example, just if I'm looking and this was true for faceless YouTube videos too. So even before AI, like if I'm going doing some research on some backpacking backpacks, which I am right now. You know, I'm ge g getting back into outdoor stuff and backpacking and I'm like looking at it and I'm like, if I can tell there's like a real human actually like hands on with the backpacks and he's got five of on it on his on on his lap here and he's like, Here, let me talk to you about this one. I used it like if that's not that, then I quick I skip it. 'Cause I know it's like a it's chat GPT is just kinda quickly writing a script. Or even if it's a human, it's just kind of taking everybody else's reviews and then like turning into a video script and then kind of reviewing the gear. But I don't actually know what that person's opinion is. I it's just regurgitated stuff. It's not helpful. And I think it's that way with AI a lot. so I think there's like degrees of whether AI content is good or whether it's not. There's another one though that I thought was really helpful. his name's Matt gosh, I'm blanking on his last name. Matt Wolf. Right, big YouTuber on AI, probably the biggest YouTuber that covers AI. And he has a website called Future Tools. And he's got specifically a news feed section that I pay attention to 'cause I don't I just like going there 'cause I find that he does a good job of curating what is gonna matter more. And then I pick out usually I would before doing the the bot bro segment of this show, if I was covering news, I just go to future tools and pick out the few ones that were like the most applicable to marketers. I'd also look for like X feed my X feed and stuff, but His newsfeed is completely air AI curated. Don't care. Whatever he's doing with AI to make that website looks good, works good. Do not care at all that he's using AI as that tool. It works and serves my purpose. I mean, are you learning a lot from ChatGPT?
Ken Freire: One hundred percent. So that's where may maybe the you is like what's the prag 'cause in some ways people could be listening to this and it could be totally pragmatic. Right? But what's the what's the line that you're using to be like, this is resourceful, like, as long as it's a resource for practical things is fine, but when it needs high trust, that's when I want a person.
Dan Sanchez: Yeah. A lot of people are. So the word that I'm using in order to decipher and what I'm talking to even my the founders of the company I work for and the leadership team and trying to help decide is what's the most considerate, which is the lot which is a word that marketers need to be asking. In fact, it actually I actually got this frame or reference from one of my favorite marketers. Probably my second most favorite marketing book is called Your Music and Your People by Derek Sivers. Fantastic book. And this is actually an appeal to people who are romantic about their craft, about why their craft is extended through the art of marketing. And he says it is the most considerate thing you can do as and he's speak he's kind of coming from a background of having been and dealt with musicians who are highly skeptical of corporate marketers and the man. he's like, marketing and communication is just an extension of your art and your music. So breathe the brand, the feeling and the emotion of the music and pull that into your marketing. And that's the most considerate thing you can do for your fans. We have to be thinking about our audience. Think about what they're going through and ask, like, is what we're putting out on social the way we're doing it with AI the more considerate or less considerate? Are we just doing it to save time? Or are we doing it because this is actually adding more value to them in a way that's really helpful? I had to weigh this myself, because I discovered with codecs, I actually need to do a step by step. I'll I'll record an episode and do a step by step on how I did this with codex. but I discovered, I'm like, like I have the ability to create infographics with codecs and ChatGPT's image generator. I'm like, I have a bunch of I have a whole site full of a thousand recipes. It would be really difficult for me to do this manually, but if I give it an image of the what the food looks like, all the ingredients. It can do like a flat lay image of like the plate and little arrows pointing to all the different ingredients of it. It's kind of an interesting image. Is it more considerate to have it on the post? Yeah, because I'm showing what all the ingredients look like outside of it. And it kind of creates another visualing way to make the recipe attractive and maybe get someone over the edge and trying said recipe. Right. I already have a real image of it, a real human shot image of the food. So that's not fake, but the top lay infographic of what's in this is another way to visually see at like really fast without reading the ingredients list what's what goes into this recipe. Right? All the different whole foods and ingredients and stuff. So is it considerate to add that into every single recipe page? Yeah. Is it considerate to then take that and post it on social and link back to a recipe? And none of these recipes are AI generated. They're all like human made recipes. Yeah, I think it is more considerate. It's another way for people to discover really good and healthy and tasty food.
Ken Freire: So so with that, you know, you had mentioned the the the picture, the original picture of the of the plate was a real one. Is it considerate to make that a f like a AI generated one? To make it a little bit more polished? You know, 'cause we've before even AI we've done that. We do that with real estate, we do that with any picture, we've we make it, we clean it up. Why not just do it with AI?
Dan Sanchez: I don't know. I could but I find that the photographer has done a good enough job that I don't really need to. And I just find that it'd be one of those things where AI because AI can have variances of going off, I'd rather just leave it the way it is, you know? for now anyway. And I think some people will questi because this AI thing will become more and more out of control, I'd love to be able to say all the pictures are not done t touched with AI at all, or at least these these specific lead images are not touched with AI at all. To be able to say that sometimes I think is helpful for credibility's sake. this to be able to say like none of our people are AI people and our images, so that when they look at our testimonial images, you know, the the before and after pictures of weight loss and stuff, they can trust them all. To say like we never use AI people. Like even in our advertisements, even if it's a like maybe an appropriate place to do it, we never do it. So you can trust that our testimonials are all real. You know, so like there's you have to take trades with these things. Now are there places where AI people work? Like with my personal brand I do it all the time, but I'm also the AI driven marketer. So of course I'm gonna keep doing thumbnails of me being AI generated. It's a more appropriate place.
Ken Freire: Yeah. So it sounds like in in order to know the demarcation, it's being considerate. I hear credibility a little bit there like where like it's gotta what parts is it credible? High trust is needed in those scenarios. because man, I I'm thinking about it to me, sometimes I feel a little subjective, honestly. If I'm being real real honest, where it's like This moment, I'm gonna use AI and I'm okay with it. You know, like for example, medical advice. Sometimes I ask AI some medical advice, and I'm like, hey, what do you think I should do about this? And I'm like, Do you really know? Is this really trustworthy? Maybe I should go to a doctor. But then I go to a doctor and I'm like, Do you actually know what you're talking about? Information, like, so there's an interesting wrestle with credibility and trust even with people as experts today.
Dan Sanchez: For sure.
Ken Freire: At least for me.
Dan Sanchez: Yeah. Yeah, for sure. I think we're all using AI and talking to experts is kind of what's going on right now. Every doctor knows they're being like they have to know they're being evaluated against Chat GPT every time. I'm using it. I think it's one of the most common uses for AI right now is medical stuff. so that's why ChatGPT is kinda dabbling with like uploading your health information into it. I wish I can get in the beta of that 'cause I'm already using a project to like track that journey. I'm gonna be do I'd probably do even more codecs soon. but that is that is becoming a big thing. so I think it matters for every single company. I think the leaders need to get together with every company. Marketing leaders, CEO, founders need to get together and wrestle with where do we go, where do we not go? What's the approval process for this? For another example we had is like I was using a ton of AI images in Facebook ads. And when we looked at them, they're like, actually, these are fine. I'm like, good, because they're making a lot of money right now. and they were like, no people. It was like a picture of the product, a supplement pack. package. And then we were using some kind of like background image of where it was sitting to like over exaggerate the features or the outcome of it. So if you have a product called Beauty Sleep and it's like something you take before you go to bed, it'd be like sitting in a dreams a cloudscape, very dreamy, magical, like you're just gonna have a magical night asleep, you know? But we're used to seeing images like that in ads, right? Before AI, we just did them in Photoshop. If you had enough budget to afford a designer that could manipulate images and come up with these crazy dreamscapes like that, right? But we're used to seeing it, so we don't really care whether it's Photoshop or AI. It does not matter. Even if you're even if the audience is sensitive to AI, it still doesn't matter because we're used to seeing over exaggerated product benefits played out in illustration, Photoshop illustrations. So if it's AI doing it, it doesn't matter.
Ken Freire: Do do you think that you know that that same process it's been years now that we've been over exaggerating on that with product products? Do you think if in with AI, if people keep over exaggerating, AI fake people, like all these different things, that at some point it's gonna get normalized as well?
Dan Sanchez: Yeah. It could, but right now it's not. I think there's something different to me where I draw the line ethically is when you're trying to I I think it's actually they're just lying. If you're using videos of fake user generated content, like an influencer or some girl sitting on her couch with the beauty sleep product, be like, I just tried this and I've I as soon as I took it, I got drowsy and fell asleep. That's essentially like a lie because you're trying to convince them that this really happened. The point is are if they if they see it, but when they see the dreamscape image, do they know it's a lie? Is am I lying? No. They they know I'm over exaggerating the features in order to make the point. And that's okay. Like it's it's are you trying to deceive people? Is that your point? 'Cause if it is, you've crossed the line. And I think the dis the part around deception is what makes it the clear like where where you can draw a line in the sample. Like don't lie to people. And that's where I think a lot of people go off the rails in B2B all the time with their email outreach. Hey, I mean, every podcaster gets this. Hey, I've been listening to your podcast in the gym. It's like my favorite thing to work out to. You know how many I got this same line over and over again because whatever software they're using found that that was a good hook and everybody started using it. Because it sounds personal, but it ev it's a lie. It's like leading with the lie and it's it it just screams, don't trust me, every single time.
Ken Freire: that's awesome. Yeah. Well, I mean this is this also comes to marketing with false promises, right? Like how many times they promised the whole world overpromise, under deliver. It's it's still a lie, right? And and we're talking about the same thing here with AI. Like so many people are overpromising with AI now and it's just like, like you can't deliver. A few years ago, we we heard people talking about agents. my god, we're gonna have agents, it's gonna change things. we still don't have like a fully functional agent that I know of today that's like crushing it.
Dan Sanchez: Getting close.
Ken Freire: It's getting close. I th I think by the end of the year we'll have something, or maybe but by the first half of twenty twenty seven. I mean, I'm even playing around with some stuff, but I'm Ooh, this could be it. But it's not there. It's still gonna be version one, right? Version two.
Dan Sanchez: Yeah, yeah. Yeah. It'll just keep getting better and better. where where are you running into AI slop and wrestling with this?
Ken Freire: I I have found more and more when it comes to write like copywriting, the AI slap is so prevalent and I don't think people recognize it. even my wife, like, you know, she'll she'll be reading something or she'll read it to me, she'll be watching a reel, and I'm like, that's AI Slop. And she's like, What are you talking about? I'm like, that script is completely written by AI, like they didn't even edit it. I I've actually come to a place now where I was using a lot of AI to write my stuff. and I've kind of toned back quite a bit and just like, I'm just writing it myself. I'm gonna copy like it feels more relational. There's just a way that I write that AI hasn't fully been able to mimic.
Dan Sanchez: Yeah. Yeah. Yeah. Same.
Ken Freire: And I'm not saying I'm the best writer, but I think at least for myself and my audience it feels more authentic and I actually get more reach when I do it. I can tell. It it just helps better.
Dan Sanchez: Mm. Yeah, yeah. It's interesting 'cause we're in a place, a day and age, where a lot of people still can't tell what AI writing is. but more and more people are waking up to it. I think like LinkedIn is a really good example of like a place where everybody's very, very like savvy when it comes to spotting AI writing. because we've they've been talking about it for well over a year now. It's probably been like eighteen months at least where they've become hypersensitive to AI content.
Ken Freire: Yeah. I for aware.
Dan Sanchez: And I used to use AI content, but I was like feeding it like we've I've like I've talked about on the show multiple times, like the my process for it and the even the custom GPT I've given away. I don't use it anymore. And I don't think you're not using it either. We're just writing now on LinkedIn because a because the crowd has gotten so sensitive to it. And I found myself getting sensitive to it. So that's probably worth unpacking. Like what's changed in us, knowing that we were doing it before and even reading it before and it was fine, but now we've become like Like as soon as I could tell the post is written by AI, I kinda skip it. And I know people skip it when I do it too. So I've just actually had more fun just like riffing and just writing ugly content sometimes, knowing that it's probably gonna get better b better response because it's not AI. It's kind of a weird place we've come to with LinkedIn.
Ken Freire: No. Well, I I think part of it is just the the for at least for me, the creativity of writing. I I felt like I was losing that muscle of like, how do I write my own hook? How do I write my own progression? And then also my vulnerability aspect. Like every time I would write something that I I'm like, I think this is vulnerable for me and my audience, AI would just change it and tweak it and I'm like, No, no, no, you're you're hiding the vulnerability. You're actually not accentuating it. You think you are, but you're not. Or other one other things that I found is just like nuances. It's it's Not good with specific nuance when it comes to certain words I use. again, coming from a faith-based background, I'm like very particular about certain things. and then I f I ri follow a lot of marketers on LinkedIn, and I could tell the ones who aren't using AI to write it, it's just written so well. It's so like the emotions are there, the the the flow works really well. And there's another book that you recommended a long time ago to me, Dan, Marketer as Philosopher, right? You remember that? That's the number one book.
Dan Sanchez: That's the number one book. If this wasn't if y Music is your people is number two, Marketer Philosopher is number one.
Ken Freire: Yeah, that that blew my mind because the one of the main principles of like how do you get people to micro yeses or micro line of thinking? And one of the best marketers when it comes to copywriting is that they're really good at like, here's point one and here's the next logical point that I'm trying to make. AI doesn't do a really good job at that. It actually goes point one to point four. You know, and you're just like, wait minute, how did you get how did you missed a few steps? and So that's why I went I've gotten back to just writing it on my own. Sometimes I'll go I'll throw it in to edit and go back and forth, but I just realized, man, it that's just taking too much time now. I'd rather just write it. If it's good, great. If it's not, like I'm still practicing.
Dan Sanchez: And at the same time, it's kinda like on LinkedIn, you're trying to compete for attention and you're trying to stand out. And when it gets flooded with AI content, what stands out? Non-AI content. So there's that part, maybe that's what we're intuitively kind of going after. It's like, well, we're trying to get attention on here. So the way to the way to get attention is to stand out. We're not gonna stand out with AI content when everybody sounds fairly polished. Cause even when it does write good and it does link you through the thoughts and it does though with a good rhythm and it even communicates your point correctly. you could still tell it's AI most of the time and you you just lose the credibility. So if you want to stand out, don't use AI content. So I'm like, all right, fine, then we'll go back to writing it dirty.
Ken Freire: Yeah. And and I will say this. I I'm there I'm writing more and more dirty. There's still sometimes I write posts that are, you know, AI driven very much. Like when I post my podcast episodes real quickly, I'm just like kind of like, here's a quick rundown of it. Because one, I recorded that a podcast episode three months ago, so I don't remember. So like AI is that's where I found like AIs helpful. And that's why I said it it feels sometimes a little subjective. But I'm like, hey, here's the things I learned, here's what I said during the top the The podcast, go check it out. and that's a real quick, like it's almost like a news flash newsletter, you know, just like, hey, update.
Dan Sanchez: I kinda wonder where it's gonna go in the future because there are times where I do like AI content. for example, my newsletter is generally all AI content when I've done it, I actually fallen out of rhythm with doing it again, even with the help of AI, because it's so much copying and pasting. I think I found a way to like fully automate it so once an episode drops, Codex can take it, then upload a full H fully done HTML draft and then send it. So I'm still working on that. But at the same time it's kinda like, well, like Yes, AI's taking my podcast transcript and turning it in. So it's like all unique content based on things actually said on the show. and the way it's considerate is that some people like to read and not listen. I've had a number of people know that I'm a podcaster be like, I just don't like podcasts. And I'm like, that's fair. Like, cool. You don't like listening to content, you'd rather read content. Got it. I'm the opposite, and that's fine. Like we just have preferences. Cool. but if they wanted to get my content, the only way to get it before was either watching it or listening to it. So by putting it into a newsletter format is considerate because now I'm giving them a practical way to get the highlights of the episode without having to watch or listen to it. that they would have had to either have taken the transcript somewhere and found it and then run it through their own AI. So I'm just saving them a step or a couple of steps if it's landing in their inbox or whatever. Is it slop? No. It's it's being considerate. It would it be better if I hand wrote it? Yeah. Do I have time to handwrite it? No. So is this the best option? Is the most considerate thing I can do? Yes. That's that's kind of like where I think it is where it's like it really depends on the situation, but you have to ask, is it considerate? Is are you lying to people? Like these are all the kinds of questions you have to ask in order to figure out for you, p your personal brand or your company, like should we be using AI or doing this by hand? Or a combination of both. Certainly you should be having a human in the loop and checking it before it goes out. but There's still places where you can have a huge advantage. Like if you don't do use AI at all and you go all manual, that's certainly a a thing you could do. I I think that's a disadvantage for most. I if you want to go all in on like we're only human run kind of a thing. there I mean there's there's companies that only do a close, you know, like artisan wool jackets and they only do hand hand spun with the old school tools from two hundred years ago in Italy and they charged premium dollars. So like there's always there will always be a place for all human everything. But they become very, very niche things. So it's not for everybody. It's for very few people can get away with we're just gonna do everything manually the old school way. Even music and all this stuff. There's still people that like Daft Punk did that whole album the old school way. Like what was it? I for can't remember that that out they had a really that that disco album they did was like done the old school d the old school way, not with all the digital tools. It was done analog. and it was cool. It was great. But again, like most people don't do it that way.
Ken Freire: And and and I think going back to marketing, right? There's gonna be but what is marketing ultimately trying to do? Trying to communicate their brand awareness. So they use a lot of writing, they use a lot of images, they l use a lot of video. So when you think about those contexts, it's like, how can you start to still use AI as an assist, right? Like sometimes I throw stuff in there to assist me and be like, I like this verbiage. I like that, I like how you said this sentence. I'm gonna take that and now I'm gonna run with it. I still do that, but and I think that's where the I think a lot of people are gonna go and I encourage a lot of people to go. Like use it as an editor if you don't have an editor. but don't let it write for you. I just found that it's just not been helpful at the end of the day.
Dan Sanchez: Yeah, more and more people are waking up to the fact that what AI sounds like and they can discern it and they're losing credibility over it. That's the one thing we're toning down to trim healthy mama the most is AI writing. And we actually have a more strict approval process for that now before we publish anything. AI images has gotten off main mostly because it's making money. So if you're at a point where you have some leaders and they're very doubtful or scared of AI, one of my favorite things to do as a marketer, and this has been true for a long time, anytime you have like a well, we shouldn't do it that way. Well, why? Well, because I don't feel like it works. Be like, cool. Well, would you like to know for sure? We can run a split test. You know. Something with like Facebook ads or landing pages. You can just split test image A over image B or this headline or that headline. It's been the it's been my ultimate c like card as a marketer for a long time. Usually I I vet like executives' bad ideas through split testing. And you know, sometimes I'm proved wrong and it's amazing. And then I move forward with their bad idea. But it's a good idea because you know what? That's what's made more money. so when I told the one of the founders of the company, particularly the one that's like the most in charge of finance, I was like, Well, the AI images are working really well. Like they're actually getting like eight X, ten X, sometimes thirteen X, ROAS on Facebook ads. He's like, Ho wait, what? yeah, we should probably keep doing this. I was like, interesting how fast that turned.
Ken Freire: That's where pragmatism wins the day, right?
Dan Sanchez: Yeah, yeah. like, wait, wait, wait, wait. That's making money? Yeah, I don't care. I don't care what the other founders say. We're doing that. So it's I have four founders to kind of juggle. It's great. but like they all they all came around. Like once we had ac actually had a conversation of like what we liked or what we didn't like and what we can accept or not accept, we actually came together with like a I probably still need to draft up a document that would get more clear over time. And honestly, it's one of those things where it needs to be an ongoing conversation because people's how they perceive things changes quite rapidly in this society. Like perception of AI is changing. what AI can do is changing. How we use it, how effective it is for our marketing teams is changing. How our our own internal capabilities, how we use it changes. So the it's shifting all the time. And I just think this whole AI slop versus useful AI content just needs to be an ongoing conversation.
Ken Freire: Yep. Hey Dan, I'm gonna put you on the spot for people who are listening to this. They may be like, ooh, when you write that document, can I get a copy of it? do you wanna be on the hook to share that with people?
Dan Sanchez: Yeah, I don't I don't think that'd be a problem. I think 'cause if anything, it'll be a net positive for the brand to even have them consider like why we're doing it. And for the people who are like maybe I g gosh, can you believe it? We ac like Trim Healthy Mama speaks to like obviously mothers or women in general, kind of like between ages of like generally thirty eight to sixty five is kind of the range. obviously there's younger and older and and even even men that take the product sometimes. But like There's overlap between our audiences. I've I've met a few where they're listening to both the show, their show and my show. so for those of you who are the Trim Healthy Mod mama audience that tune into this show, the reason why I could justify using AI as a marketing, like cause some because some of the audience who might be anti-AI in general, like they might be anti-social media and they're anti-AI, I don't know. might say, like, why use it at all? And the reason. I would justify using it is because it's more cost effective to use it in a lot of cases. And if I could be more cost effective, I don't have to raise the prices of the products as much because I can get more marketing spend or do more for marketing with a lot less effort and a lot less time. So that brings our cost down, which makes it so we don't have to raise the prices as fast for our products in order to have healthy enough margins that the business can actually sustain growth. So That's like my one appeal to like users is like a you want people using AI because everybody wants more higher quality for less prices, faster, right? Like everybody wants that. at the same time, we also want to emphasize and really value what is distinctly human and what is really important. So that's also something I'm navigating with the brand is how do we lean into more people? So even for that brand, I'm leaning into bringing in more creators in house and hiring Either as a contract as contractors or bringing them on full time, I'll be looking into like bringing up more creators than just Pearl and Serene who are kind of the faces of the company. Because I want to double down on human faces, human personalities, human stories, and the real stuff. And we use AI to speed up everything that's not as essential for humans to do, but want to spend more time on the stuff that only humans can do. which is actually like talk about real. human transformation, which AI can only like guess. So yeah.
Ken Freire: Yeah, it can mimic it. But even it mimicking it, it's just like it's kinda like playing t the game, what was it, telephone? Like you ever draw something, and then you're trying to get everybody else to draw it. It it kinda that's how I see AI a lot. It's like you need a really good assistant, but even then after a while it it can just veer off.
Dan Sanchez: So I will publish that and let people see it and just to kind of watch how the process is evolving for us at the company. it may be help be helpful for kicking off your own conversations. but this will be a work in progress. Like we just covered it two weeks ago, and I'll be kind of giving updates on that conversation as we go along because I think this is gonna continue to change and evolve, like I mentioned. But let's wrap this up. I wanna know. Was this a helpful episode? Would love to hear from you guys. Reach out to Can or I on LinkedIn. That's where the kind of the best place to reach out. shoot us DMs. And we'd love to hear from you guys.