Attention Marketers: You Are Trapped in an AI Bubble

(upbeat music) - Welcome to Marketing Brew Weekly Morning Brew Inc's podcast for the Marketing Industry. I'm Kelsey Sutton. - I'm Katie Hicks. - And I'm Jenny Mywin. - So something I've been thinking about a lot, YouTube, is that consumers or brands are not really on the same page about AI. And it's got me thinking, I think marketers might be in a little bit of an AI bubble. - Hmm, same more. What does this bubble mean? - So I recently covered some data about consumer perceptions about AI and its use in advertising. And the results are not exactly amazing for brands who are sort of using it every single day, every way they possibly can imagine. So one of the things, I have more data to share, but more than three quarters of consumers believe AI makes ads feel less authentic. - Three quarters, that's 75% of consumers. - Synthetic media feels inauthentic. - Right. - You cut up gas. - I feel like that's such a biting piece of data because authenticity is like what every marketer is craving and working and striving for, right? - So much of buzzwords. - Yeah. - Yes, I totally agree. And I think basically the long and short of it is that consumers largely do not like to see AI in their advertising. They do not respond well to it. They don't find it trustworthy. And some of them say they're less likely to buy from brands who are using it in visible ways. Meanwhile, on the other side of the equation, AI is becoming more and more and more of a topic. It's more ubiquitous than ever, or particularly on marketing teams who are trying to figure out how to use it, what's the best way to use it. So my question for you is will brands just continue pushing AI at all costs and hope consumers come around that they decide to change their minds about this or is there a white space for brands or potentially agencies who embrace a no AI or a less AI angle in marketing and their processes? But first. - Are you out of breath trying to keep up with AI in the changes it's constantly creating? - It can be exhausting, we know. That's why Morning Brew created a new podcast with PWC called the Intelligence Shift. - This eight episode series focuses on how AI is changing and disrupting different industries. Join host Dan Priest to discuss real stories and lessons learned from AI with expert guests. - For example, did you know companies achieving the biggest productivity gains are actually boosting wages and headcount, not replacing jobs? - And jobs that give AI complex work are falling behind, whereas jobs that give AI the basic work and give humans the more expert tasks are thriving. - If you're interested in deep dives on those kinds of insights, this show is for you. - Listen to the Intelligence Shift wherever you get your podcasts. - Okay, I wanna dive into this Harris poll report, it tells you that you reported on. So what were some of the main findings besides the fact that two thirds of people don't like AI in their ads? - Yes, so yes, it was an exclusive report from the Harris poll, the forays and infillian, they worked on this together. What they found is that I'm just gonna rattle off a couple of stats and we can get onto it. More than two thirds of global consumers view AI as largely a marketing ploy. So when they see AI used in marketing copy or in other instances, they are like, no. 78% of consumers believe AI makes ads feel less authentic, which I said at the top of the show, three quarters of respondents said they find brands to be cringy when overusing AI. And around two thirds, 63% of consumers said they would be less likely to purchase from a brand that uses AI generated ads. - I mean, it's pretty, it's gonna go away. - Yeah, I mean, I think one of the things that I wanted to highlight is this cringiness, this feeling of, I'm less likely to actually purchase, I mean, those to me indicate like real business problems. We talk about authenticity, that's something marketers are always chasing, they're always looking to achieve that, but also they want results, they want people to like feel affinity for their brands and ultimately purchase from the brands. And to me, I feel like, I don't wanna say warning bells, like siren alert, but I do feel like this is such at odds with what we're hearing. I mean, I'd love to know your guys' thoughts on this. - I mean, another stat that stood out to me from the report was that more than 54%, I was 54% more than half of people agreed that they hear so much about AI that it's starting to annoy them. So it's like, it's to the point now where everyone's talking about AI so much that it's just like enough, like people are getting a little tired of just like every brand touting how much AI they're using or using it and not disclosing it and just feeling like, what can I trust? Why is every brand only talking about AI? What else is there? - I think that particular sentiment comes from though, there is this kind of like, budding of heads between what consumers are saying, they find cringey or annoying versus like, what people in the industry in the sea suite are kind of expecting from their companies as well, because we were just like pan-lion and like obviously everyone was talking about AI all the time. And I actually remember just watching like, like social video from a like B2B content creator before we went and was like, if you're going to can like, make sure you have your AI hat on. Like I like normally like don't consume AI but I know that everyone will be talking about it and this kind of like girl boss kind of way. So it's like, you don't want to stick out in this like corporate crowd to be like, oh, I actually don't like AI or even just like poking holes at how AI should be used is not necessarily the common tactic amongst people in the industry, but at the same time, we're getting the very opposite from people who are on the consumer side and who are actually seeing some of that work. So that is tough to navigate. - And I think that's what Kelsey who mentioned earlier with like the bubble. It's like it's the small talk, you know, topic de jour and everybody's talking about it. And it's like if you say something critical, it's like, well, you're not on the cutting edge of this or like you think this isn't good, like this is where the industry is heading. But it reminds me so much of a few years ago when everyone's talking about the metaverse and it's like, we all hopped on the metaverse train and then look how that ended up. Like that, like the medicine's so much money on it that just ended up going up and flame. - Yeah, yeah. And I think that's the thing. I mean, I like the idea of exploring this bubble a little bit more because I think marketers are marketing two different groups. And by that, I mean, if you are a marketing professional, if you were an executive, I think internally inside your company for job purposes, you need to be understanding of AI tools, what the possibilities are, whether that's generative AI or other, you know, - Fishing skills. - Yeah, right, exactly. Like there's so many tools here. And I think part of the new table stakes is you have to be coming like with ideas and insights about it. And then internally, there's all this conversation about efficiency and how quickly can we do things and scale up creativity and all of this. But then we know that the consumer largely, again, based on a lot of this, they don't find it to be innovative. Like they don't see AI and they think, oh, innovation, ooh, et cetera. They're thinking like we're tired of this, we're sick of hearing about this, we're less likely to trust. There was another stat that said 73% of consumers said they would be less likely to trust an ad they suspected was made using AI. That's not even, that's not even that it sure was made with AI. That's, I think that AI was involved and therefore I feel. And again, I think that's such a disconnect. And then also with those brands who are maybe indicating to their investors, they want investors to know that they're embracing AI. And like, that's-- - Well, investors love that too. - Exactly. - And they're on the forefront of a big trend. So they get pay off by saying that. But yeah, if your customers aren't liking what you're putting out, if they find it distrustful, then that doesn't achieve the means that you've been trying to achieve. - I think the suspicion point is so interesting too, because it also speaks to how ubiquitous the AI conversation is because it's not even that a company has to be going out and explicitly saying that they used AI on their ad creative or whatever it is that the consumer is actually interacting with. But it could be someone just like on their morning commute, waiting for the train and seeing something that looks like a little bit off. And it could just be like weird creative, but they are now associating it with AI. And if they have a perception around the concept of AI, there is a lot more negativity that can come with that rather than just being like, "Oh, like that ad didn't really hit for me, like moving on, whatever." - I've seen people talk about this, like to extend it to the entertainment world. I've seen people talk about this with like the New Avengers Doomsday, trailer people are like, "This looks like AI Slop." I don't think it was AI. I think it was probably CGI. I'm not totally sure. But the fact that people are now like, if they don't like how something looks calling it Slop regardless whether or not it is AI or they can confirm that it's AI or not, is not a good sign, I would say, for generative AI. - Yeah. - Using a greater purpose. - Well, and I think that's such an interesting point, Katie, 'cause I think we're seeing that just, like that's a great recent example. And I think I've seen that at a couple other times where there's like a weird CGI. There was this moment where people are doing like virtual IRL. Do you remember what this were brands? We're like, "Oh, we're like wrapping a subway card "it looked like a lipstick tube or something." - Oh, but it's all CGI, right? Like it's all, but again, people see that and it's like, "AI is becoming this catch-all for like ugly "or something that's not real." Or something that they realize is intended to look real, but isn't. And I think that is like the warning sign, is that culturally it's becoming this blanket term for like bad slash, "I don't like this," or "I'm being tricked in some way." And I think when brands are thinking about building trust, Like, you know, if you go and when you put out an ad and you're like, this was AI generated, et cetera. Like it's not doing to the consumer what maybe that intention is. And I think that's a real challenge. I mean, I, we've seen some AI generated ads. I mean, what if you guys thought about this? I mean, in some ways, maybe this is a long time coming. Maybe this is just data that's proving what we've kind of seen, I don't know, anecdotally. Yeah. How do you actually think a founder who left Big Tech to start a men's grooming brand is doing revenue wise? Is it good? Yes, you can actually see how Amazon ads helped crowd skin scale to over $500,000 in monthly revenue on Amazon ads rising stars series. This stocky series follows small business owners as they build their brands from the ground up, face challenges and scale. And it's not like they had huge budgets either. These are small brands trying to make it happen with help from Amazon ads. Watch it now at advertising.amazon.com/rising-stars. That's advertising.amazon.com/rising-stars. Well, I know, Jenima, you wrote about proof of reality and some of the ads where they show behind the scenes. I feel like this is happening at a time where people are also liking the Odyssey just came out. And people are liking the practical effects of it all and seeing how this Cyclops was actually real. I think people are interested in this whole behind the scenes idea of how they build that, how they do that Apple did this. And they showed how they made their new logo a few months ago. I don't know, it's interesting. I think people are starting to value the creative process a little bit more and brands that can tap into that might be rewarded. Yeah, it's interesting that Apple, example in particular because I feel like I can't think of pre-the AI conversation. How or why Apple might have put out something similar being like, here's how we made this because I think a lot of the appeal of a brand like Apple was that slickness, that sleek, modern image that it's just already beautiful and packaged and presented to you in this way. But a lot of those adjectives being sleek and modern and presentable is really easily created with AI now. So it's like they had to peel back the layers and bring people in. And there was a really big positive reaction around that. And people are interested in it. And it kind of creates this whole other content form for them, of being like, oh, you can get to know us as a brand more. And also this is an entirely different type of advertisement that we haven't done before. So it is an interesting opposite reaction to the general AI boom that we are seeing. It's also interesting because I feel like it creates an ad with legs. You have your ad that you actually made with practical effects or manually or whatever. And then you can do your behind the scenes as follow of that. It's a two and one. Yeah. People are tuning in. They want to see how did you do that. At the same time, we also know that that's very expensive. So Apple has lots of resources. So they can spend that time. They can film all of that content ultimately, right, from doing practical work. And I wonder when there's a brand who says, I have a really small budget. I'm trying to stretch it as much as possible. AI generated assets are a great way to fill out a social calendar or whatever. I wonder if this-- I wonder how this affects on-- how you think about this on different platforms. Because if I'm looking at an Instagram ad, I spend what? Two seconds looking at it. Max, you see it, you look at it, you scroll. You're not necessarily lingering on it being like, is this AI is this real, et cetera? And so I kind of wonder if part of this is way-- because I think about Coca-Cola had their big splashy holiday ad that was generated with AI. And that was extremely controversial. And also, Svedka for the Super Bowl did an AI generated or AI supported Super Bowl ad. And so they used that to animate the robots that dance for their first ever Super Bowl ad. And we also know brands are sometimes kind of using AI to make fun of AI. And that feels a little bit like a winking to the, we know you don't trust AI. Neither do we. But they're still using it. But yeah, we're using it. Yeah. Right. Probably to show to their investors or whoever. Like, look, we're using it. We know how to use it. But we're also tapping into this consumer sentiment. I will say that both kind of uses of that in both of those examples. Like, it generates conversation, right? Whether you are like, I'm in on the joke and I'm using AI, but I don't like it. Just like the consumers don't like it. Like, you get the benefit of people being like, oh, that was-- like, they get it. They're thinking the same thing as me. And they're showing like kind of like the opposite side of the AI conversation. And then you feel aligned with your customer. The Coca-Cola and Svedka ones, I think, are really interesting to me because I feel like tapping into the conversation around AI, like really is like a brand awareness generator. But I wouldn't say that Coca-Cola and Svedka are necessarily brands that need a ton of brand awareness. Like, I feel like the conversation around the Coke ad in particular was so largely negative. And there's so much goodwill, typically, held towards the holiday campaign in particular. There's a lot of happy, cheery vibes. I love those polar bears. Yeah, people really do. That's a brand-ass that the people are like, I'm excited to see it in the holiday season. So I think it was an interesting move to kind of lean so hard into AI and really make it known. Like, it wasn't like they were very covertly using it or just subtly using it. It was like, no, this looks significantly different from all of the different, the past years. I felt to me kind of forced consent a little bit where it was like, you know, like I remember one of the Coke executives came out and was like, this is the future. Get on board or don't. Like paraphrasing. But it was something like, this is what's coming. And like, I don't think customers respond well to that. Where it's like, you better like this. They're like, well, we don't. Like, what are you going to do about it? Well, and I think again, I think that's such a, it's a tricky place to be in because consumers, I do think there's this huge effort to make AI ubiquitous in so many facets of life. And now we're seeing consumers saying, we don't like this. And there's a lot of data to back that up. And I think the question is like, do you just, or you're like, well, the consumer is wrong. I'm not going to do that. And I do think that there is an opportunity because if the data shows us that consumers do not feel good about this and in different applications, the brands that say, I'm not doing this, I think that there's an opportunity to generate a lot of goodwill. They stand out. But the question is, how do you do that, honestly, because there's a lot of AI use on the backend, sort of in the shadows. I don't want to say, like, just not visible to consumers. So Netflix just came out and it said that they used AI kind of unknowingly in like many different titles. I don't know if it's not announced, but it was like, yeah, I think when consumers don't realize its AI, they are less likely to complain. I think right now we're seeing kind of the early iterations of generative AI. So it's more visible that the Svetka ad. I think people could clock immediately. They're like, this is a little uncanny valley. There's something here that feels a little off. So it felt like this is probably AI and then Svetka confirmed. And I think they also were like, we want to be one of the first brands to have our advertisers the Super Bowl using AI. And like, look at us for such an innovator. But yeah, I think it's still, I wonder like, you know, what the Netflix example like as it gets better and people have a harder time identifying AI generated content, you know, what does that look like? - Yeah, or does that cast again, cast a big cloud over anything that seems off, even if it's just some bad CGI or something, you know what I mean? Like I really do wonder about that. Something I wonder about is, you know, using AI sort of in the idea generation sort of concept stage. So according to this same report, more than half of industry professionals say they're using AI for concepting, idea generation, about half said they used it for visual asset creation. And then about four intents said they use AI for processes like copywriting or audience targeting and personalization, media planning and placement. Those to me feel like much less visible to the consumer. You're not going to necessarily say, this ad was concepted using AI and then filmed like with real people. But I do wonder about what this means on sort of the creative side of things because if you are an agency and you're like, we used AI to idea generate and concept, well yeah, couldn't any other agency also use the same AI tools to your in-house team? Like why do you think it's in the first place? - Right, exactly. And I think it's almost like devaluing like the purpose of doing it in the first place. I also think some of the best creative campaigns to me are like the weirdest. And as we know AI is like generates sort of in the middle, right? It kind of is an average-- - It's homogenous on the internet. - Yeah, it is an averaging tool most generally. And so I just really wonder about what that looks like on the agency side. If you're trying to stand out, you're trying to say we're an innovator, we're cutting edge. But now everybody has access to the same tools as you, like that no longer is a differentiator at all. - It also pulls from existing media. So the more you use AI generation to come up with your ideas, how much does it become the snake eating its tail? You know what she's like? - Right. - Suddenly it's like you're pulling from other AI generated campaigns and then it's just like one big slop campaign. - Yeah, it kind of reminds me of I feel like people will like go on like a image generator AI and like feed it into itself to see what happens and it gets weird and like not in a good way. And I wonder if that is something that is happening on a much larger scale that not only will people be able to clock that it's AI faster, but they'll be like it's getting worse. Like there's, it's instead of more AI usage honing the tool so that it's better. If it is only feeding on existing AI material, there is no new idea coming into that system to kind of like make it better and make it. interesting. Also something I think that's interesting about going back to this FedCAD. I think Kelsey, you said that there was, or Katie, it was something about like wanting to be the first in the Super Bowl to use AI to generate their ad or to advertise with it. And I think that there's such an interesting kind of dichotomy between being the first to use AI versus like using AI well, right? Because if you use AI first, like you're kind of putting your brand on the chopping block to be like, oh, well, this is like not the best AI could be yet, but we, but it's like we're here. We're at the starting line before anyone else's, which generates its own kind of media. But if you wait until it's better, perhaps people will accept it more. But where do you come in? That's a tough choice as a brand to make. Yeah. Yeah. Did you guys see anything about the REI ads on meta platforms? Oh, yeah. AI generated images. I didn't see this, but I did see that meta's AI tools were kind of messing up brands at. Yes. That's like that's exactly what it is. So REI ran, you know, an ad campaign and one of the ads, it was AI generated was a bike with two sets of handlebars. So people were like, and it wasn't a tandem. Yeah. And people were like, where do you sit on this bike? Why is REI selling this? They said, this is not intentional. Like this was turned on without our knowledge, without our consent, because meta has these tools to generate ad creative to support this. And Jasmine has actually for our team has also written about this. Some brands are struggling to find out the toggles to be able to actually turn this off. So this is not the first time this has happened. I think that's a really interesting element here, because the brand is like the brand name. It's not meta who's in hot water. You know what I mean? Like, sorry, that's probably like, oh my gosh, meta has, it's like you, the responsibility kind of lands on the brand. And again, I think that just indicates to me that the speed at which this is being deployed, like, maybe there's a moment to think, what is this? How are consumers going to receive this? Like, do they feel good about this? And what happens if something gets through and we don't catch it? Like, that creates a huge headache for brand affinity and, you know, just consumer perceptions around brands. Yeah, it's an interesting like piece of like the efficiency conversation too, because like something like this example, if a brand is just classically creating their creative and putting it on a meta platform, it creates a new workflow where they have to make sure that they check that the AI tools are turned off or like if it is actually hard to find, like, there are a bunch of loopholes I imagine you have to jump through, like, whether that's like in the actual software or you're like the representative that you are working with, like that creates such a longer process than just creating your assets and putting it on the platform, which is the opposite of what I feel like even AI purists are like, no, it's like really just for efficiency, like even unintentionally it can kind of create the opposite effect. Yeah, well, I think even the Coca-Cola ad from the holidays, they think they came out and said that they had to go through it like multiple times. It wasn't like they just say generate a holiday ad for Coke. It was like they went through like, I think upwards of like a hundred times, just like regenerating things, redoing stuff, like editing it here and there to make sure it got to that point. So then it, yeah, raises the question of like, how efficient was it in the end to use AI? Yeah, especially if people look at that and go, wow, I hated that more than I've hated any other, like it's just it's all that for me. Yeah. So I mean, so we've kind of talked about this all in the sense of, you know, consumers thinking something's AI generated or AI was involved in some way and feeling very distrustful. We're reaching a point, it's very early, but there's this movement from some states to require disclosures of AI use in advertising in certain cases. So in New York, they're requiring that you label if there are AI generated models or characters. So like if you have an AI generated person or an AI generated hand or foot in like a shoe ad or something, you have to disclose that. And I wonder what that's going to look like because now if you're a consumer and you say, I really don't like AI generated ads, I don't trust them. And then you see a little disclosure that says this was generated by AI. That's legally mandated. How does that change the calculus for brands? Like it's right now the idea of just we just do it and hope nobody notices or does that extra disclosure prompt a new sort of creative thing? I'd love to know your guys' thoughts here. Yeah, I think it's interesting because there's no one way that the New York state law requires a brand to disclose how to how they use AI. Like there's no official wording or like where you need to put it on the ad or where. And I think that's interesting because brands might be very sly about it and keep it small and like very subtle text where it's like, this was a generated okay, anyways, and I'm at like moving on. Yeah, moving on. I think that could potentially make it so that the effects are negligible. But if they do have to display it in a very visible way and it does end up being visible, I do wonder if that'll discourage some brands from using AI. It reminds me of even just like classic like working with creators and like the FTC laws that do require like ad disclosure. So many people have to add in a corner after 45 periods. Exactly. Yeah. 15 other hashtags buried at the very end or something. So I think that like disclosure laws have been around for some type of advertising like since the dawn of time. But people will try to get around it, whether that's on a smaller scale or a larger scale. And I do wonder if like the platform matters, right? Because it's like social media platforms, a lot of them do currently have disclosure rules right now. But a lot of people are not opting in or if it's an automatic thing, it doesn't work all the time. So it's really just like how well will it actually be upheld and how well will people or how much will brands and advertisers or creators like actually uphold these these rules. And I don't know. I feel like we are still struggling with the like ad disclosure on a basic level, especially on social media. So I can't imagine that just like having a law roll out will like fundamentally change like exactly how it actually plays out. But I do think that when you see it, it like it is kind of double edge sort of like, Oh, like if I have a bad perception of AI, then it's like, Oh, I might have a bad perception of this ad. But am I at least grateful that they didn't try to like pull one over on me and like told me from the top that this is what I was looking at. And do I care if that is the case? Because what if it actually looks good or it makes me feel good or it got the information across in an efficient way. So I think there's a lot of layers to how a consumer will interact with a disclosure. Yeah. And I wonder if the disclosure actually helps support that trust. And we see that that, you know, that person in shifts. But, you know, for now, I do think it's important to pay attention to how consumers say they react to these sorts of things because it's it's definitely not not as easy as just saying new tech lets do it and everyone's going to love it. Yeah. Yeah. Well, thank you so much for joining us today. If you have thoughts about consumers thoughts on AI use and the ways brands are using it, email us at [email protected], which you can also find in the show notes and catch you next week.