Introduction
**Episode Description:**
Hey, have you ever wondered if a podcast could thrive without real voices behind the mic? 🤔
In this episode of *Content Universe*, we’re diving deep into that intriguing question!
Join Mikkel Svold, the brain behind Montanus, as he shares his wild experience with an AI-powered podcast engine.
Is authenticity being compromised? Do listeners even care if the conversation is completely generated by AI? 🎙️
Let’s explore the exciting possibilities and some tricky pitfalls of AI in the podcasting realm. With Mikkel’s personal stories and sharp insights, you’ll rethink the crucial role of the human touch in content creation.
Curious about what the future holds for storytelling?
Don’t miss this engaging conversation about technology and authenticity. Tune in now and get ready to share your thoughts on this AI-driven revolution! 🚀
What You’ll Learn
1. How AI reshapes traditional podcasting experiences.
2. The potential of AI to mimic human conversation.
3. Drawbacks of automating podcasts with AI.
4. Branding implications of AI-generated content.
5. Importance of validating AI-produced content.
Episode Transcript (AI Generated)
Hello and welcome to the Content Universe. Oh, my god guys. Today I’m a little bit blown away. Today, one of my colleagues introduced to me an artificial podcast engine, meaning something where you can take a bunch of input, feed it into an AI, and it will create a conversation, a podcast conversation between two people about that particular topic. And I’m baffled. I’m like, wow, this is new ground for AI in my opinion. Because when I listened to this podcast, the podcast example, the AI-created one, I must say it was very good. It was like very, very good. And you know, for those of you who know me, you know that I’ve been saying all along, I believe in the person to person, human to human conversational podcast. I believe in that as a piece of content. But now I must say, I’m a little bit in doubt, and that is what we’re talking about today. For those of you again who don’t know me I am Mikkel Svold and I’m the content, no I’m not, I’m not the content. I produce content, we produce content, and I’m the founder and CEO of Montanus which is a content company where we’ve produced content for high knowledge companies, so that could be engineers, it could be some sort of advisors, consultancies, those kinds of people who have something complex they kind of want to have people understand. And we do podcasts we make blog posts, we write social media posts, We only do the content stuff. So this whole AI era is really something that is digging into everything that we create and we use it a lot we use a lot of AI because not using AI just to me seems a little bit well not an efficient and if and also not an effective way not just efficiency but also effectiveness, you can be so much more effective and also efficient obviously when you use AI. As I said we’re going to talk about this little thing called AI podcast because from what I heard today I was really before today I was pretty convinced that AIs in podcasts would not be a thing because I was convinced that people would want to listen to people. People would want to listen to human beings because it is the human beings that we kind of can connect with. We can connect with the stories, we can connect with the backgrounds also. We can also connect with the examples and experiences they have. Today I really had to question whether that is a.. whether that is the truth or not and.. and that’s actually just what I want to talk about today because today is not going to be one of those how you do it kind of podcast episodes for me. It’s going to be more discussion on what.. what matters in content, what matters in particular in podcasting. Now, the first question that I want to pose because of today.. the first question I want to pose to you guys is will people care? I thought… I was pretty convinced people would care people will care that it is human beings behind the microphone they will care that it is genuine or maybe not genuine that’s a wrong word I can get my… in Danish you would say my ears in the machine I don’t know with that but I’ll get in trouble from from AI experts or from AI people but will people care, will listeners care whether it is AI-generated voices that sound like 100% human-like or will they care more about or will it mean something that it is real people talking to each other. I’m not sure especially after today because one of the things that I really enjoy when I’m listening to a dialogue podcast a discussion podcast is that people tend to go out a tangent, they tend to give examples of past experiences they’ve had, they also tend to talk in a very human way so that could be interrupting themselves with new points or with new ideas that just pop-up with new examples and also having, of course, all of the speaking mistakes, small jargon stuff that you would say, linguistic details that, you know, linguists will really be on about this. But this is what makes speech human and what I kind of felt today was that this AI-generated voice discussion did a lot of that, maybe not deviating too far from the… from the root that you were settled, but it did deviate. Now, I just heard a podcast episode from Lex Fridman yesterday where he interviewed Donald Trump. And that, ladies and gentlemen, is probably deviating too much from the question because, quite literally, Mr. Trump did not answer a single question that Lex Fridman asked. He just answered some completely different thing. And that’s probably too much at least for my, for my taste. But will people care? That’s a really big question. And then also, I think, there’s another question. And that is, what drawbacks are there, if you sit down, or if you just set up a system, where an AI can basically shoot out podcasts on behalf of you? Well, there are a few drawbacks. Obviously, you’d have a knowledge drawback. So you would have a peer review knowledge drawback. And what I mean by that is that when you put an AI to create the to do the talking, you don’t know what it’s saying. Of course, there’ll be instances where you can probably edit like wrong things out. But as of now, the software didn’t seem to allow to for editing. And also, what was really clever about the software it was called notebook LM. What was really clever about the software is that was that you can you put in your own sources, so you kind of load the AI with the sources that you want the podcast to be based upon. And you can load up to, I think was 50 different sources. So that could be there’ll be journal articles, right that could be peer review journal articles, or, or only stuff on a specific topic. And that will all then go into this machine. But you don’t have exact control of what’s said. So if you are say a brand that is pushing out a podcast using this technology you kind of do want to control, what is said, and you also kind of do want to control the manner of which it is set in. And then of course I think if you are a corporation what signals are you sending that you use an AI to do your podcast? Is that a good or bad signal? Because what you can question of course as a client, as a listener, you can question whether the knowledge that you are sharing in the podcast is actually knowledge that exists within the company. Right, so that is also a drawback right there. And then of course there’s a big drawback not just because you don’t know what it is saying you also don’t it’s really hard to verify whether it’s true or not because at least that will mean that you need to listen through it and you also need to read all the articles that you’ve fed into the AI in the first place. I think this new technology I think it somehow will require a form of validation or form of where does this come from you need to be kind of transparent with whether you using AI to produce a podcast or not just simply because it sounds so human. It sounds so real and just like you would with with the images often you would you would have like the bigger magazines the bigger newspapers they also they all have like these little texts that AI generated image if they use an image that is AI generated and can we always count on people putting on labels when they AI generate stuff no obviously not because in many cases you would want to conceal the fact that it’s an AI who’ve done it but what I think is going to happen and what I hope is going to happen is there will come some kind of validation on whether it’s a human or a real thing let’s call it real or an artificial thing as if we put those opposite of each other right so there’ll be some kind of validation and one of my colleagues and another colleague actually he we talked about it and he said probably it’s going to be some kind of blockchain technology so it could be some not a proof of work like but like a proof of proof of state kind of Bitcoin and let’s not go into that but a Bitcoin a proof-of-state what’s called blockchain that’s sorry but some kind of technology that is well it’s not possible to mend with you’d have to have a secure technology could be blockchain technology that will validate and that it will tell the consumer whether this is AI generated or not because that is I think is going to be important and it’s something that I hope that legislation will will think really hard about and I know they are and be also kind of quick about because all the technologies already existing right now and now that kind of comes back to whether or not people will care whether or not people will think that it is important to know if it’s AI generated podcast episode if it’s AI generated discussion because basically this is just two AI agents talking to each other that you’re listening to which is insanely crazy I can’t I can’t get my head around it anyway I think this dilemma is something that we’ll have to deal with in content production for the next many years and it is something that we’ll need to deal with on all levels of content production not just the actual production but also on a strategic level and probably also on a business unit level business and on a business scale level and of course on a governmental level and basically a worldwide level the UN should probably do something or whatever I think that is going to be this Content Universe episode for today so I’m just baffled about this podcasting taken over by AI I’m I’m all confused I’m all I’m all excited I can’t I can’t decide whether I’m afraid or excited I think it’s a good mix of both anyway that is it for today’s episode and I really hope you enjoyed it if you did give it a like give it a share and of course hit that subscribe button and bell notification button because then you’ll get a little little ping every time I post another episode like this one and of course share this episode I think this episode is actually and I’m kind of you know padding my own back here but I think this episode is worth sharing maybe not because of what I said but because sharing the idea about validating and and what what value human input has I think that discussion is just so important to have so share with someone you think it’s really relevant for and have the discussions out there I think and I would really really love to hear from you because I’m just I’m just really curious what do you think about this what do you think about AI taking over more and more interactive stuff you know play pretend that it’s a human being talking to another human being what does that this mean to you guys let me know on email you can find me at podcast at montanus CEO and that that email goes straight into my own inbox or or you can of course find me on social media I’m probably easiest to find on LinkedIn where you can just search my name Mikkel Svold and I’ll pop right up hit me up with that I’m really curious to know and yeah see you on another planet in the content universe
