Why the AI Rollout Is Such a Mess (and Where Humans Still Win) with Sean Richards | Concordia University

Sean Richards 0:00
If something is just a a a simplified uh prompt, then you're gonna know within reason, okay, I think that is probably AI, but the more thoughtful uh your prompt writing becomes or the the the inputs become, then the more indistinguishable the output becomes. And you're like, I don't even know this has to be human because it had a very human touch. And I think that's what people are missing in all this. And I know with creatives, the moment you bring up AI, people start to clutch their pearls. But I do think that there is something to be said about the creative process is still at play. You just have to think about differently.

Kevin Kerner 0:37
Hello, everyone. This is Tech Marketing Rewired from Mighty and True, and I'm your host, Kevin Kerner. This new season, we're focusing on what we call past the pilot, what marketers are doing now that the AI experiments are over. For the opener, I wanted someone who teaches this stuff and actually uses it. Sean Richards is an educator at Concordia University and runs Creative for a Living. Of course, he also teaches young students about the future of AI. He is also writing a musical about Bigfoot, and we get into that and a bunch of other things like why the future of our educational system needs a change, why the AI rollout is such a mess, and where humans still win. So let's get into it. This is a fun one. This is tech marketing rewired.

Sean Richards 1:27
Sean, great to have you on the show. Good to be here, Kevin. Thanks for having me.

Kevin Kerner 1:31
Yeah, thanks so much. This will be our first uh first episode of the season, so I'm really excited to have you on. Let's just get it kicked off. Um, you have a really interesting position now as an educator, uh, and you I know you have a creative background too. Um, and I I wonder if you could just tell us a little bit how you got started in education. And then I would be really interested to see like how you see the shift now that we're entering the agentic era. You've been an educator for six years now. You got a really interesting background, and and the timing is of what's happening is kind of crazy. So just give us a little of your background.

Sean Richards 2:02
Yeah, absolutely. Uh I've been a creative uh for the past 10, 11 years, copywriter by trade, creative director, uh done a lot in that space. And becoming uh professor was a happy accident because I'm probably I was the worst student of all time in college. You know, D for diploma, uh A for effort. And I'm right with you there. You get it, you get it. And I just wasn't very interested in the way a lot of education was set up, but I had some incredible professors along the way who showed me, oh, there actually is kind of a cool path to create, and it turns out it was the old school Socratic method. And so um one of those educators was the dean of the School of Business and Communication that I studied at at Concord University, and she, I guess her better judgment, decided to hire me after I got my MBA. And I came in, I taught a capstone for marketing, absolutely loved it. And uh my whole focus was how do I create a class that was angled around what I wish I was taught of just the actual because a lot of it is just so textbook anecdotal of like, hey, how to know if you're ready for market or whatever, and just things that no one's ever gonna ask you. Like I taught a sales course that was like, well, uh, you know, let's understand what the manufacturer is and all this. I'm like, no, like that none of that matters. Do you understand a pipeline? Do you understand how to make a cold call? Do you understand you know funnel management? And so I just kind of took that sense of, man, I I got my butt kicked for the first couple years outside of graduation. How can I set students up so they're gonna hit the ground running day one? And that's kind of what led me on this path, and now we're at almost eight years as an adjunct professor of marketing communication and uh some sales classes here and there. But it's been a wild ride, and AI has been just kerosene on it, and it's it's been really fun to see. I I do not have the boogeyman mindset that I think a lot of academics do have with AI right now. You know, I I think that it is obviously an incredible innovation, and as well, I'm sure we'll get into, there's probably some nuance that needs to happen in all this too. But it's been a fun front row see seeing with the students, especially with Gen Z.

Kevin Kerner 4:14
Yeah, and I wonder what they uh what you hear from them. I just read a uh stat this morning that that adults under age 30 are more likely to say that AI will have a negative effect on society and on them personally than those that are older. What do you hear from the students right now about the use of AI, getting into the marketplace at this time?

Sean Richards 4:32
Aaron Powell Well, it's it's hard not if you're under the age of 30, it's hard to not be a doomer if you're scrolling TikTok or Instagram when so much of what's being reinforced. And and to be fair, this is a tough market right now. If you're a creative, you know, let's just call a spade a spade. I think what we're seeing right now is a correction in the market from the overhiring that happened during COVID. And I think that you have a lot of companies that uh are over their skis and they're using AI as the convenient excuse to reduce headcount. And I get it, but also it's tough when you know you just had so many people telling you, hey, this economy is just gonna keep going, going, going, and the correction hasn't really happened the way that it needs to. I'm no economist, but the way that it in theory needs to. And I think unfortunately, Gen Z is holding the bag right now. And so they're seeing, like, ah man, what is the point if this thing is literally being trained to replace me? I don't think that's actually the case. I think that we've seen the last three months the tea leaves have kind of played out a little bit, and I think a lot of these companies have scaled back. I mean, you look at Uber blew through their AI budget in the f in Q1, and I think a lot of people are saying, like, oh, this is not tenable. We can't afford to use Claude code on every single ticket that comes our way, and so now we have to scale back a little bit. And so I I do believe that it's gonna come back around, and in the next five years, there's gonna be a little bit more optimism around our relationship with AI. But understandably, if you're a college student right now, just the way that things have played out the last three years, I think you have every right to be frustrated. And I don't think that it is indicative of where we're going, but it's something that I try to be mindful of when I'm having these conversations with them because they can't get away from it. It is it's inevitable. You and I both know AI isn't going anywhere. Now, I don't think it's going to be the the foregone conclusion that you know the Altmans and all of them have been touting, but I do think that it's it's like the dot-com. All the everyone and their mom was getting VC funding, and then when the when the tides came in, you had a few players left. And I think that's what we're gonna see in the next two to three years. After all these dot AIs get washed away and the jobs kind of kind of settle, you're gonna see the quiet rehiring happening. People are gonna get uh lower position levels at a lower price point, and they're gonna quietly get Gen Z back in the workplace.

Kevin Kerner 6:57
Yeah, I think your comment on tokens is really really the thing because like when Fable came out, uh it was interesting that it came out with uh um free time up, I think it was gonna be up to the 22nd when you could use it for quote unquote free. And then it would be token. You couldn't you couldn't run it on your subscription anymore. So that token subsidization that all these uh LLM companies are going through is gonna end, and we're gonna have to pay through tokens, and that's gonna be a lot more expensive. I heard another, there's another research company. They don't actually tell you. I have the $200 like max plan on Claude Cut Claude, and um, they don't tell you how many what that actually costs in tokens. Like what do you what would you have to pay instead of the uh for your token usage? And there's a comp research company that said it was about, if you max that 200 subscription out, it's about $8,000 a month worth of tokens. And there's just no way. So there's a subsidation phase that's happening right now, and I I think it's real. I never thought of it that way.

Sean Richards 7:55
I mean, at that point that's fewer lower devs.

Kevin Kerner 7:58
Yeah, that's right. And you can't just do there's an unlimited backlog right now because you can just do everything. And once that settles, um, it'd be really interesting to see how that affects the the youth job market because you're gonna have to go back to having them use the these agentic tools, but not like not like build all the things, get rid of all the people, because that's just not gonna happen. Um what are you um how are you uh changing the way you teach them because of this now? Or are you? Or how or how do you how do you um approach AI in the classroom now?

Sean Richards 8:34
Yeah, well the first thing I tell students is whether you realize or not, you've been using AI for most of your life, and for Gen Z especially. And it's just the the packaging looks different. It's more comprehensive than it's ever been, obviously. But I mean if you if you've used Grammarly, if you've used Auto Tune, if you've used Siri, if you've like it's been in like we've been inundated with it for a long time, and I think we've just never seen the marketing of AI so front and center. Uh and so I I try to break that. But the other thing too is is I'm really you know the old teach a man to fish and leave for a lifetime adage. But uh with with students, like I I think GLP ones are a great analog here where we've got this miracle drug, right? And we don't really fully know the effects of it and what's gonna do. But if you were to sit here and ask me, hey, Sean, is you know, uh a weight loss drug that we're not fully cognizant of better than chronic heart disease, yeah, of course. And that's that's a trade I would take every day. However, if you don't fix the habits, if you don't teach good discipline and good fundamentals, then a shortcut doesn't work unless you know the path forward and or the the actual path to get there. And I think the same with AI, it's a wonderful innovation, but innovation without discipline is ultimately useless and and dangerous. And what you have now is if we don't address this in the classroom and teach them, hey, you're only as good as your prompt writing, you're only as good as the thought uh as the inputs you give AI, then you're just gonna see more and more slop in the market. And you know who really gets gets stuck with all this is the poor middle managers who have to read the slot coming from their top level and the slop coming from their their directs, and you know, and they have to sift through all this where everyone else is just saying, uh, give me the path of least resistance, just write this brief or whatever. And it just becomes this mess of everyone's trying to outclaw each other. And so my goal is yeah, use AI on every assignment I give you, but use a thoughtful approach to this and ask questions of, hey, what are you trying to solve with AI? Like, what would actually maximize this? And sometimes I think a lot of times my students find like, oh, I could have written that by myself and have a more thoughtful answer faster than the time it took me to actually go in and re and correct the inputs with AI and tweak it. And sometimes that's the case, not always, but just getting them to think critically has been the number one goal here. And then once they inject AI in that, then it's just they're off to the races. And that's it's you it's doing exactly what it should be doing.

Kevin Kerner 11:07
Yeah. I wish do you think that C chain, do you think that change in how higher education is treating AI is is gonna happen? Because because Mike, I have kids, three kids in college right now, and they're um you know, it kind of runs the gamut. They have some teachers that are that are say, look, you need to learn the context, but you can use the heck out of it. If you don't learn anything from my class and you're just using AI, you're gonna go on to the next class and you're not gonna do very well, like the next level of classes. So uh and then there's some teachers who are just completely against any use in any field. Do you think that like will that change in higher education over time?

Sean Richards 11:45
I I think higher education is due for reckoning one way or another. And I mean the system is just so I mean, a lot of universities operate as though it's still 1954 and that this job market is just waiting for their bre best and brightest to come out of here. When the reality of it is, if you're a business major, uh unfortunately, you'd probably be better off just graduating high school and starting and failing a business than you know getting a four-year degree with you know what, sixty to a hundred thousand dollars of debt, uh depending on your circumstance. Now, that is a broad stroke, and obviously what please don't hear me say I don't think that further education is helpful. I absolutely do. And I think it's a matter of you know, if if you go into that, like what can I glean? How can I improve here? And you know, especially if if you're in a more quantitative degree or you're looking for something like going in the med the medical route or legal route, then yeah, obviously you have to go through those credentials. But I think especially what you're seeing in the business world is uh a need for academia to realize, hey, this is just a bad business model that we've haven't we haven't kept up with the rising costs of of living, the rising costs of of uh inflation and and all that's that's packaging around us. And so we kind of have to give them the tools to be dangerous and hireable right now, as opposed to four years from now when they graduate and they have their cap and gown. And so I think there's a lot of folks who are just gonna get weeded out uh because of that and already starting are in a lot of universities are coming around to this, and I think you're seeing more certs being offered. But the hard thing about AI certs is it's moving so fast that by the time you're you know quad competent or you you have understanding of of design, you've already gotten passed up if if you know within three months, and so you have to learn the next you know iteration. But I I think I I really do believe that uh universities are getting leaner. I think that they are are starting to focus on the actual like, hey, let's offer more nursing degrees, let's offer more you know uh quantitative courses here, and I think you're gonna trades, yeah. Trades. And I've been a big proponent of, you know, there's I mean, this was uh a stat as of you know 2012, so take it with a huge crate of salt, but back in 2012, there are 6.5 million trade jobs available in our country. I think that has leveled out a little bit, but there's still uh millions of available jobs uh that I think universities need to figure out how do we better partner with you know two-year accredited programs to offer other solutions and other alternatives, and then how do we take that same approach of apprenticeship to mastership in business? And I think, you know, I don't know what the exact answer to that question is, but I do think it looks something like having someone who has vetted an experience in the field come in and say, similar to like a board certified doctor, hey, this group of legitimate doctors says that these candidates are ready to go. I think you need to start seeing that across the board in business as well. Now, who holds that title? I don't know. You honestly, you'd be better than I would for that, Kevin, with your pedigree, but I do think that we need to get back to apprenticeship in some capacity, and that's where universities have have failed.

Kevin Kerner 14:57
Wow, yeah, yeah. Well, I hope I I I hope I certainly hope we have more educators like you and voices like that, because I I couldn't agree more, especially with four more kids getting ready to go into higher education. I'm like, I've something's got I'm hoping something changes, both pricing and approach and the amount of time it takes. You know, there's there's so much there that that needs to be fixed. Your perspective as as a creative director and copywriter and educator and academic and someone that actually works on the stuff is super interesting to me. And it kind of goes into the next um conversation I wanted to have with you around like the I think last year, maybe last couple years have been a lot of pilot and experimentation using this stuff. And there's pressure, at least, whether this is real or not, to like go beyond that pilot phase and actually get to the production phase of things. Like we're done with the pilot. I think a lot of that's coming from boards and senior executives and you know, the pressure down on the employees. I wonder if do you think we're beyond the pilot in the rollout of all this stuff? And and you know, what can actually be operationalized beyond experiments at this point?

Sean Richards 16:06
Good question. I think we're expected to be beyond the pilot if you ask any you know C-level executive. I think the reality is a lot of employees still don't necessarily know their left hand from their right in terms of a critical way to use it. And I'll I'll give you an example of you know, uh going back to the slop. Uh the I had a client who, you know, leading up to the World Cup, they want to just flood the internet with SEO of trying to, hey, we're a travel brand and we want to catch all the folks coming in to America to experience this. And so effectively they asked, you know, my team and I to write you know what turned out to be about 30 articles in the span of a week. We're like, this is just it's like you know, a three-man operation. Like at pre-AI, it would take me like eight to twelve hours to write a really thoughtful research article. Even with AI, it cuts that down. But I mean, 30 in a week is just insane. But those are the kind of demands that AI has created is you know, the fallacy was, oh, it's gonna give you more time to be creative. Like, no, we all know it just gives you more time to create more slop and more like quicker outputs. And I think that's what a lot of employees are feeling is they don't necessarily feel like they have the autonomy, the bandwidth, the safety and job security to actually learn this thoughtfully. Now, with that, I want to give huge props to Canva. I don't know if you saw this, but uh a few weeks back they announced that they basically cleared the decks for all of their employees across the company and gave them a week to do nothing but learn good AI hygiene, good AI practice. And I believe that is the gold standard. And for everyone who's just you know cynically looking at like, well, it's Canva, like, well, I think any company with enough prep time, you can figure out how to make this an actual, you know, everyone's got their AI czar or whatever, but it's always turns into they don't know what they're doing either. And you you've just kind of said, yeah, go figure it out because that's what our investors are asking for. But to actually sit down and say, well, again, what is the problem we're actually trying to solve? How is AI going to improve? Is it, hey, we have way too many health desk tickets and we don't have to fix this? Is it a matter of we're not keeping up with AEO and we want better online searchability and presence? Great. Take the scientific method, take whatever your hypothesis is, and then train around that and then A-B test it. And if it doesn't work, you can tweak it and tailor it. But right now, what you have is is so many folks are just said being told be an end-to-end AI expert without real any kind of guideline or or guide rail, guardrail, and as a result, like you just get more and more slop and more and more confusion and more and more frustration. Either people are getting laid off or they're getting pushed out of their jobs or they're getting burnt out. And I think to fix that, you gotta just basically do what Canva did and say, all right, let's clear the decks, let's sit down and say what is the why, and then let's build from that.

Kevin Kerner 18:54
Yeah, to your um to your 30 articles in one, you know, in a week sort of thing. My experience is that it is that maybe we're a little past the pilot in the understanding of how to use the tools and the harness. You know, you can you kind of understand how all the pieces fit together now. And then um, but there's still like to be able to do that at scale, do something like that, you have to build some infrastructure and skills around the what you're trying to do, such that the output is just completely locked in. And at that point, it could even be built into an app, and then that app can do the thing. But we are not at the stage at this point where the the productionized AI can happen. Uh and certainly an a an employee can't do that on their own. You mentioned something about autonomy. Like part of the issue with our clients is that they they want to do this stuff, but they just don't have the keys, they don't have the ability to do it. So it's like they're told go do the AI thing. Oh, but uh you can't we're not gonna give you access to all the things and you can't build anything. It's it's a lot easier for me as a business owner and a small business to get in and experiment, but to operationalize that at scale, it's just I think it's tougher. And we're not out of the pilot phase yet. Um have you seen any particular ways that you've studied that is an optimal way to roll these things out? I mean, I love the I love the Canva idea, but do you have any other examples of successes or ways that you know this can this can be made better?

Sean Richards 20:30
Uh yeah, I mean adopted. Sure. Anecdotally, I have found a lot of success in following folks on Twitter who are like a video standpoint.

Kevin Kerner 20:44
You know what I mean? Like Yeah, and Substack's good too.

Sean Richards 20:46
Substack's awesome and Reddit, and it's the the wonderful double-edged sword of the internet is it's undefeated. And you're it you're just gonna get such a conglomerate of good ideas, and they're probably not coming from open AI, they're probably not coming from Claude, they're probably coming from the practitioners, and you're just some kid who is just goofing around with VO and figured out how to make a better, you know, Marvel set, you know, set piece or whatever. And that's like, you know, I I kind of started playing around with that and figuring out prompt writing on and some video work. And uh that's kind of where I've I've gotten my best. Like, oh, cool. Like if you write it in in this code, it's you're gonna get a little bit more of a clear understanding. And so that's what I would say is like just see examples that people are doing, like, that looks pretty sweet. How do they do it? And a lot of people are sharing the recipes, like no one's gatekeeping around here. And I I think with as a result, like you can kind of bring that back to your creative team. Now, you got to be careful because if you get a uh you get a CMO in there, like, well, great, I don't need a videographer anymore. And that's I think a lot of where a lot of the the uh anxiety kind of starts to to fester. Yeah, but the reality is a good CMO looks at it and says, Okay, great, we can integrate. This with our videographer and make them that much more effective.

Kevin Kerner 22:04
It's dangerous. Like you get the you get uh you get the client saying that I can build all these in lovable. I mean it's just so easy to do. I saw some guy do it and it's just like it's not necessarily not that easy. I think that's a really good point, is that um where I learn the most and I maybe mature the most is wanting to study the stuff from the thought leaders and not from the brands themselves. Um like I follow the AI Daily Brief podcast. It's it's really good. Like it's just only 30 minutes, but it keeps you up to date on all the things and it gives me inspiration every day. I think you have to be a seeker of that type of information.

Sean Richards 22:41
Totally.

Kevin Kerner 22:42
And um man, to get uh taking it back to your students, um uh do you find that any of them are have the time or want to dig into those sources? I mean, if I was in school, I don't I don't know back when I was in school, I probably would have just been playing drums or something. But uh do you find that that do you tell your students like you gotta go look for this stuff and do they do it?

Sean Richards 23:05
Some. I I think it's you know, I I'd I'd say you probably get five to fifteen percent that are industrious enough to to want to add on. And that's you know, and that that probably was not dissimilar to how it was 20 years ago, you know, in terms of whoever like there's always gonna be a minority of folks who are trying to learn a better strategy and a better way forward, and the majority are gonna be later adopters. And I mean that is human nature, but with students, I think the the important thing is it's kind of like talking to my my my kids, and I've got an almost three-year-old and I've got a one and a half year old, and we're really big on you can do hard things and kind of implementing in their their their brains early on, like you're saying you can't, but I know you can, and I I want to push you to to try this. And as you see with your kids, you've got eight of them. You can see that the more and more they take small steps and they get little wins, the more and more confidence they have to say, I can take on bigger challenges and I think I can be successful. And it's trying to do that because I think so much of not to like again roast higher education, but a lot of it is predicate around let's let's cattle call all these students into one room and let's just teach them the the path of least resistance, how to be the most average version of an employee you can be, be hireable, and that's it. But there's not a lot of hey, you can challenge this and you should challenge this, and you can build something better than what this Fortune 500 company has because they've just kind of settled the path of least resistance as well. And I think for me as a as an educator, where I find the most pride and joy is when I see my students start to click and they realize, like, oh, it is it is easier than I thought it was. And it's just a matter of I didn't see that that was a possibility. And like for me, I remember in college the thought of of selling a company just felt impossible. And then I saw one of my best friends do it, and I realized, oh, my buddy is not, I mean, he's he's a smart guy, but he's not the sharpest knife in the drawer. He's smarter than I am. But if he can do it, then sure, it's obtainable. And I think that's what a lot of students need is just saying, like, oh, it is doable, and then they can they can handle that. But I don't think we we give them great examples right now across the board.

Kevin Kerner 25:15
Yeah, maybe what they don't see too, and if any of your students watch this, that um what they don't see now are the conversations I'm having with CMOs where they say, I can't get my I can't get my team to adopt this stuff. And they're starting to what I tell them, the successful ones that I see are hiring uh people right out of school that are have been AI curious, that have actually gone and done something in AI. They might just know how to prompt, or they could be, might have taken a Python course or a you know, just a simple coding course, and they've played around with it. And there's there's opportunity for them to be hired as the AI, junior AI and automation engineer at a company.

Sean Richards 25:55
Yeah.

Kevin Kerner 25:55
Marketing teams are all looking for how do I get my team to do this stuff? And what I tell the senior executives is hire someone right out of school. They don't have to have a CS background. It's probably better that they have a marketing background, but someone that has a marketing background that studied this stuff, just something, and give them the tools and just let them go because they'll learn it. And that it's true. I've seen it happen several times. Well, I hired my son here from um he went to Franciscan University and he got a biology degree, but he took a Python course. And then he so he understands it. And so, you know, that it's a huge opportunity for younger people too.

Sean Richards 26:30
Um I think to that point too, Kevin, it's like the the thing that doesn't necessarily transfer on a resume is grit. And when you see a a young person that has a a desire and a drive to figure something out in a good attitude, that to me is more valuable than any pedigree. And a lot of times, what I love about teaching at Concordia University, and it's a great school, and there's great students and there's great instructors and and you know, admin, but we don't have the pedigree of a University of Texas or you know, it it fill in the blank. And in some ways, I think that's a benefit because students aren't coming in with like, oh, I am McComb's school of business. Not that students are, like UT has great students, obviously, but I've taught a lot of students or I've hired a lot of students from UT, AM, whatever, and you kind of have they have to unlearn a lot and they have to become adaptable and flexible versus when you're in an environment where maybe you you kind of have a little bit of a chip on your shoulder or you have a little bit of that Tom Brady. I was not the first round draft pick, I was actually a later round draft pick mindset. It makes you more adaptable and teachable and coachable to make you that much better out the gate. And so I do think that we should focus on finding those diamonds in the rough and empower them. I mean, to your son's point, it's like he had a biology degree, didn't necessarily translate, but he figured it out because at the end of the day, he had the DNA to to be curious, and that's really what you want. Not to quote Ted Latho, but that's that's the secret ingredient.

Kevin Kerner 27:58
Yeah, 100%. Yeah, and I've met some of the Concordia students, they're great. Well, Audrey's fantastic, who you know who we both know. So uh we high or uh she's working for us now. So something about those students there. I I completely agree. Um now when you look at um human judgment on on this work versus what you can automate, um, and you're you know you're also a creative, so you see some of the stuff. Where do you think that human judgment still wins that you can't automate? You can't you just can't do certain things.

Sean Richards 28:30
Let's use Suno as an example. Have you played around Suno much at all?

Kevin Kerner 28:34
Yeah, yeah.

Sean Richards 28:35
It's so fun. It's it's a really fun. It's a marvel of a tool. And I'm I'm a songwriter, and you know, I've I have played around, and and you've heard the versions of like the like any other AI output, there is slop, and then there's stuff that's like, dang, that's really good. But what I've learned in that is Suno's gonna make you can say, hey, write me a country song about you know losing my dog in my truck or whatever, and it'll come with a you know, buy the numbers, generic, detailed earworm. Yeah, it's gonna sound like a good earworm. Yeah. But if you want to really punch up and say, like, how do I get like those lyrics that just stay with you and everything, then you gotta put a little bit more thought in the prompt itself. And like we go in and you know, my my buddy Derek and I were, you know, he's my producer on every song uh I write, and we'll just go back and forth and we'll have voice memos of like, hey, I think this is the melody for the the hook, I think this is the melody coming out of the bridge or the verse, whatever. Hey, I want this line, like, and we would just kind of tweak it over and over. And then in the process, it just becomes the Suno output becomes that much better to where you're like, oh, I really because you know, if something is just a a simplified uh prompt, then you're gonna know within reason, okay. I think that is probably AI, but the more thoughtful uh your prompt writing becomes or the the the inputs become, then the more you know indistinguishable the output becomes. And you're like, I don't even know this has to be human because it had a very human touch. And I think that's what people are missing in all this. And I know with creatives, the moment you bring up AI, people start to clutch their pearls, but I do think that there is something to be said about the creative process is still at play, you just have to think about differently. And I guarantee you, every demo that's going out in Nashville or LA or New York is using Suno or some similar tool because why wouldn't you? Like it's it's cutting down your your turnaround time, and then ultimately you're gonna get that human voice that's gonna come in on top of that. But if it was written by a human on the front end and then you use Suno, similarly to like uh writing a good article with Claude. You know, I had the idea, I had the thoughts, I had my my prompt that I wrote, and I you know articulate these are the keywords I'm trying to hit, you know, blah, blah, blah. And then it spits all that out. Then I go in and I tweak it and I massage it. I've still cut down my writing time. The input and the output still human. It's just I cut down a lot of the iteration, and that's what Suno's doing as well. And so I think that's that's kind of where we're moving and starting to see again, like we've been using AI forever, and we've been using it for iteration forever. It's just never been so robust.

Kevin Kerner 31:09
Yeah, it's even that it's even like that on the development side. Like I have to. Something is happening here. And I had to open up, oh, I know it was um, it was styling on a on a page, and I was like, well, couldn't we just write classes for these styles? And and I had to tell, so it was it was working in a C this AI was working in a CMS, and it was having a hard time like injecting what we had done into that particular CMS, and it wasn't thinking that it could just name classes inside of the CSS and then just create its own separate CSS file. And I had to tell it that. And I would that is just a basic the use of classes for CSS. It's like, what? Shouldn't you know that? And this was using Opus 4.8. Wow. Um, and it just got so wrapped around the axle it couldn't figure out the the thing. So it's that little, it's a little now at the same time, like it also wrote an app to inject all the stuff direct from HTML into the CMS, and I was just I couldn't believe I was blown away that you can run HTML direct into a heart like a really rigid CMS, and it just goes in and works perfectly. It's just it's just crazy.

Sean Richards 32:21
AI's got me back and I'm an app developer. It's got me it's got me overly confident in a lot of ways for sure.

Kevin Kerner 32:27
But it's it's amazing how you need to open up Inspect every once in a while to look at the code and saying, oh, I get it here. I did that yesterday, just in Webflow. It was like couldn't figure out this video problem, and I had to open up Inspect and just show it to them. So, oh yeah, Kevin, yeah, that's the problem was.

Sean Richards 32:40
And do you think that maybe is is the the biggest hang up right now is is with AI is that we're missing the quality assurance piece because we're assuming the trust factor with AI is, oh, it's doing all of it for me. I don't even have to really get in the weeds of this thing. And I wonder if that's where we're missing out on the best possible version of an AI.

Kevin Kerner 32:58
Yeah, it's like your Suno example, it's the same thing, it's quality quality assurance. You know what good it is, and you can and you can think of it. It's like um I also a couple years ago, I had wrote a LinkedIn post about transcribing, using AI to transcribe, and then getting a transcription of it and then post you know posting something or writing something. People did not like that at all. And I was like, well, it's just it's my thought. I'm just getting it organized, and then I tweak the thing, and now everyone does it. So it wouldn't be it wouldn't be a big deal at all. But it's that QA process that's informed by context that you just can't you can't replace. That's why your students, I think, to your point about using A in the classroom, it's like you have to teach them the context, yeah, and then you use the stuff full on. But if you don't know the context, you're just gonna become a slop creator.

Sean Richards 33:45
It it does feel like we're kind of uh playing moneyball a little bit right now. And you know, you have the example of like all the scouts back in the day were like, I just I got a vibe. I've got a vibe I'm going off of. And you know, then you're like, well, okay, but this guy gets on on base, and that's what we're trying to accomplish here. And so again, it comes down like what are you trying to solve for? And then it's just gonna take a while to coach people to think about the actual solution.

Kevin Kerner 34:11
Yeah, 100%. Cool. Well, this has been great. Um I can I could talk to you a lot longer. Uh it's a lot of fun. Re really good stuff. Uh I'll have to have you back on as we get um more input into where things are headed. Um, I do with this one thing at the end of these uh podcasts called AI Roulette, and I've loaded your LinkedIn profile into Claude, and it's gonna talk to you as the AI. This isn't me asking you a question, so it's the AI asking. So I'm gonna hit send here and see what it gives me. I know you have some very colorful things that you do, so I was I wonder if it picks this up from Claude's about to put me in my place. Yeah, that's right. Okay, it says this is from uh Claude. It says, Sean, my file flags a conflict. By day you teach you teach marketers to demand proof before they believe anything. By night, you're writing a musical about chasing Bigfoot through the woods. I checked, no verified Bigfoot, and you want him real anyway. So which Sean wins, the skeptic or the one chasing footprints?

Sean Richards 35:09
Oh man. I think uh the the winner is Sean the hypocrite. Uh I think in all that. Yeah, it's uh I I am a walking conundrum. I mean, I the the thing about Bigfoot is I will say this writing this musical, I went into that.

Kevin Kerner 35:28
You gotta talk about the musical just real quick.

Sean Richards 35:31
So my producer buddy Derek and I we're also working on a musical called Yeti or Not, and it's about a guy looking for Bigfoot. Uh it's a it's a it's a musical comedy about grief, belief, and Bigfoot. And uh we've been working on this for about three years now. We've got our first like we've we've had a couple table reads and we have a production of it coming up in in March, um, potentially one in fall, 2026. Uh but with all that, you know, in the process of writing, I did my research and I started getting into the folkloric mythic zeitgeist of Bigfoot. And maybe the skeptic in me has waned, and maybe I'm starting to actually give it some credence here. But I do think I go into most things. I mean, I don't know. It's uh I don't know if it's ADHD or I'm an Enneagram 7 or a youngest born, but I'm generally optimistic and I do try to come into this of like, hey, I believe that there's a lot of wonderful things that can happen here, but also let's just have I mean, my wing is an eight on the Enneagram, which makes me say, but but let's let's actually challenge this and let's actually see if this is resilient. So I'm a walking contradiction internally. I'm always second guessing a lot of my like you know initial thoughts just because uh I, you know, I I just I want both, I want to I want to find the truth and also I want to have a good time. And sometimes sometimes they're at odds.

Kevin Kerner 36:52
There's no way uh AI could come up with a musical about Bigfoot. There's no way. That would not have crossed the the thing. So that that is context. That's definitely the uh the definition of context. I'd like to believe so.

Sean Richards 37:06
I'd like to believe so. But AI has definitely helped us in the brainstorming of a musical about Bigfoot. I'll tell you that much. It has cut down a lot of the writing time.

Kevin Kerner 37:13
Yeah, really, really good. Uh you'll have to I'll have to post about the is the musical out?

Sean Richards 37:17
I can't remember if it's actually out yet or there's uh the full demo is out on yeti or not musical.com and people can listen to all the tracks, which don't worry, humans recorded them, human written by humans, recorded by humans. So for anyone who's gonna spiral out about that, um and uh but at the same time, I think in terms of brainstorming, uh Claude has helped me ask questions of well, how would Bigfoot answer this question?

Kevin Kerner 37:43
Oh yeah, I can't even really, really good stuff. Yeah, that's great. I'll post the link in the in the promotion that we do here. So, Sean, it's been great talking to you. I really appreciate you being on. Gavin, thanks for having me. Really, really good stuff. Keep doing that, keep keep uh pushing the university system for me.

Sean Richards 37:57
We'll we'll we'll try. Appreciate you. You need to come uh speak to the kids.

Kevin Kerner 38:00
Yeah, I would love to do that. I'd love to do that. Sounds good. Great seeing you.

Sean Richards 38:04
Have a good one.

Kevin Kerner 38:05
Bye. Bye-bye.