AI vs. Music Industry Update | Matt Connarton Unleashed
What's This About?
The conversation covers parallels with the Napster era and a question that gets personal: if you love a song, does learning it was made with AI change that? A caller also shares how AI tools can help people with disabilities turn their own lyrics into music, adding another perspective to the debate over technology and human creativity.Explore more episodes and find out how to be a guest:
MattConnartonUnleashed.com
More from Matt Connarton Unleashed
Listen, watch, and explore more episodes at MattConnartonUnleashed.com
Want to be a guest on the show? Visit MattConnartonUnleashed.com and choose Be A Guest.
Follow the show, subscribe on your favorite podcast platform, and join us live for Matt Connarton Unleashed.
Talk. Music. Entertainment. Interviews. And More.
Speaker 1: The other thing that really caught my eye because geez, you know, we talk about all this a lot on the show. Whenever we talk about music industry news, we talk about AI. We talk about streaming. We talk about piracy. Uh, we talk about this brave new world that we've entered with AI and how it impacts the music industry. And I say all the time, it reminds me, you know, for those of you who are old enough to remember, uh, This sort of seismic shift and, frankly, some of the panic around some of it that we're going through right now in the music industry and how AI impacts it.
Speaker 1: It kind of reminds me of when Napster came in and file sharing became a thing. And vis-a-vis the Internet, music effectively became free. Um, but this, this really jumped out at me again. This is from a music business worldwide music streaming fraud inquiry launched by us copyright office says here, the us copyright office has launched a public inquiry into music streaming fraud. The office published a notice of inquiry in the Federal Register on Wednesday, October 7, seeking information on how streaming fraud is carried out, its economic impact, and its relationship to AI-generated music.
Speaker 1: Now, I'm very curious to know, what exactly do they mean by streaming fraud? As I take a gulp of my water here.
Speaker 2: I apologize.
Speaker 1: Because our guest wasn't able to make it on, I have to talk more than usual, and I've been sick this week. I don't know if any of you listening have, but there's something going around in this area, at least in a pretty serious way, and a lot of people have been sick. This is the best I've felt in a couple days, but I spent a great deal of both my Thursday and Friday unconscious. I just felt terrible. I'm going to take another gulp here. So sorry about that, but I do not have a choice in the matter.
Speaker 1: It's either that or I begin hacking uncontrollably.
Speaker 2: Man. Let's see.
Speaker 1: So it says here, written comments are due by November 23rd with reply comments due by December 21. The inquiry follows a request from Scott Fitzgerald, the Republican U.S. Representative for Wisconsin's 5th District who sits on the House Judiciary Committee.
Speaker 2: His letter... to the US CEO dated May 21.
Speaker 1: Opens by asking it to, quote, examine the contributory effects of artificial intelligence in music streaming fraud and the resulting effect on the music industry. While there is nothing inherently wrong with using AI as a creative tool in music development or the organic growth of AI-generated music, specifically music, Generative AI can serve as a catalyst for criminals to perpetuate streaming fraud, unquote.
Speaker 2: So that's from the letter.
Speaker 1: He then added in the letter, quote, If done fraudulently by bad actors deploying bots or other forms of manipulation, it represents a clear theft of the royalty pool, unquote. The letter cites Deezer's finding that, and we talked about this before on the show, this number, I remember this very specifically, and it shocked me. The letter cited Deezer's finding that, quote, up to 85% of the streams generated by fully AI-generated tracks, unquote, on its platform in 2025 were fraudulent.
Speaker 2: Now, again, what do we mean by fraudulent?
Speaker 1: Because when I think of fraudulent in this context, I think of like if you're I mean, there's nothing fraudulent about just using AI to create music, right? But what has gone on is sometimes people use AI to create music and then try to pass it off as somebody else. That's a weird thing that's gone on. I remember a number of years, before this whole current AI boom even really started, we were talking on the show a few years ago about something that happened to Celine Dion, where there was this Celine Dion track circulating that apparently fooled everybody.
Speaker 1: Everybody thought it was Celine Dion, and it wasn't Celine Dion. It was completely AI, but somebody did a really good job with it. and fooled a lot of people. But I'm curious. One of the things that I think is very important as we look into this, as we think about it, as we have these discussions, is we have to very clearly define.
Speaker 2: What do we mean by fraud? What do we mean by streaming fraud? If you.
Speaker 1: Create a band that doesn't actually exist and you put out music as that band, and there have been examples of that, where the band itself is completely made up doesn't exist.
Speaker 2: These people don't exist.
Speaker 1: They're not real. Is that a form of fraud? I mean, you are kind of fooling people, but is that really fraud?
Speaker 1: And then, and then if in the process of doing that, um, you make money because you make, you earn royalties, um, from the streams, is that fraud? If you're doing it with music that people, because where it gets tricky is, well, if people like the music, they're listening to the music.
Speaker 2: Raises a question. What would you do?
Speaker 1: How would you feel if you listened to a song and you loved it? You just absolutely loved it. You thought it was so good. One of those songs you just wanted to listen to over and over.
Speaker 2: You liked it so much. And then you found out.
Speaker 1: That it was completely AI, that someone had used a program like Suno, for example. Suno is a very popular app. I've used it to create stuff. What if you found out that it was entirely made with Suno?
Speaker 2: What do you do?
Speaker 1: Does that affect how you feel about the song? Does the song now no longer mean the same thing to you? Do you no longer find it catchy and infectious if it's a song you really love? Sorry, I had to drink some more water there. I'm really struggling here. Like I said, I've been sick all week. Might have to, we'll see how much longer I can just, again, I'm talking more than I expected to because our first guest didn't make it on. But, you know, again, these are things that we all have to kind of grapple with to some degree, I guess.
Speaker 1: I mean, I can give you my answer to that, but my answer is not,
Speaker 1: you know, to the scenario that I just proposed, my answer is not something people are going to necessarily like.
Speaker 2: But if I heard a.
Speaker 1: Song that I really loved, and then all of a sudden one day I find out that it was completely AI, I don't think in my mind I could just say, oh, okay, well, I'm not going to like that song anymore. I mean, if there's something about the song that connected with me, you know, is, I mean, you know, But again, that's not going to be a popular statement. I realize that. A lot of you would say, yeah, well, if I found out it was AI-generated, I would never listen to it again. Okay, but what if it was a song that you really, really loved?
Speaker 2: You know? Anyway. I would prefer that.
Speaker 1: Then we get some little clearer definition of what we mean by fraud. We actually have a call. Let's grab this, and this might help save my voice. Hi, welcome to Matt Connors and Unleashed.
Speaker 2: Who's this?
Speaker 3: Hey, it's Jenny.
Speaker 2: Hey, what's up?
Speaker 3: I liked your question, and you might not remember, but that happened with me. Remember that song I came across the first time, Inner Fury?
Speaker 2: Oh, yeah, yeah.
Speaker 3: That was online. Yeah, I thought they were real, and I really loved the music, and I even reached out to the person who was like, hey, you know, my partner has this radio show, and we love highlighting indie artists. And then he told me the truth, and I was really disappointed because I really did believe or hope that they were real people that didn't change the fact that I liked the song. So there's one aspect of it that way.
Speaker 2: Right.
Speaker 3: You know?
Speaker 1: Right.
Speaker 3: But on another hand, I came across a song that was written by a CRPS warrior, somebody with disabilities. And they had poured their heart and soul into the lyrics for years, writing this, putting it together. And then they used technology to get their writing put to music. And it was powerful and meaningful. And I'm not going to, you know, so there's the side of technology that helps people be creative that otherwise wouldn't be able to be.
Speaker 2: Right.
Speaker 3: Does that make sense?
Speaker 2: It does. Absolutely.
Speaker 3: So I won't, like, ever just totally be like, oh, you used a computer. You stink. I would never do that to somebody. Because, you know, some of these people have poured their hearts into these lyrics. They don't have... thousands and thousands of dollars to take it to a studio and have somebody help them turn it that way or hire musicians to make it come forward. But they can use this technology to bring that feeling to life. And especially for folks who are disabled and the more technology comes out, the more they can have an equal footing in the world and create.
Speaker 3: in ways that they never had available to them, I would never just completely be negative.
Speaker 1: Yeah, I think it would be pretty absurd to say to that person, well, what you did isn't valid because you used technology to create that music that you obviously wrote the lyrics for. I don't think anyone would want to say that to that person, and I think that's a really good point.
Speaker 3: The thing is that people, like the first person that I brought that to within our own community actually did react that way.
Speaker 2: Really?
Speaker 3: And they did. One person out of many. One person. And it actually hurt me to get their response was something to the effect of, I'm not even going to listen to it. I'm not even going to give it the time of day. I'm not going to listen to it because they used AI to create it. They used AI to create the music. But every lyric they painfully wrote over years and they finally just now brought it to life and shared it. So that was my response was, you know, please remember when you're making these harsh decisions.
Speaker 3: Where people with disabilities fall into that line, because we want this world to be 100% accessible to everybody, right? Like, you want somebody who's disabled to be able to create just as somebody else does, and that might take having accessible ways to do it. It doesn't make it any less valid or cheap in it. it's still an avenue for people. And that's one thing I ask of people to think about when you're making harsh decisions on AI. AI can be good and bad, yes, but it's not all bad. And for some people in our world, especially folks among the disabled community, it is a new window to the world and getting out and locked from their homes and out into the world in ways they've never had equal footing before.
Speaker 3: So think about that before you jump on something and say, oh, I'm not even going to give it a listen. You know, maybe hear their words and then make a judgment.
Speaker 2: Yep. No, I think that's an excellent point. Well said. Absolutely. All right.
Speaker 1: Yeah.
Speaker 3: I mean... There's a lot of that going on now, and if you stop and look, if you stop and actually look for it, you'll find it. But you've got to actually go look beyond that 30 seconds and see what's behind it. What is the human behind it? What are they doing? Are they just doing something because they want to make money, or are they really trying to rip people off? Are they really trying to, like... you know, get away with something, then, you know, to heck with those people. They're no good. Right.
Speaker 3: But there is artistry out there that is beautiful and is finding a new voice in ways they never have before. And, you know, at least give them a moment. There's a new one you've heard me listening to around the house. It's a husband and wife team. They're very real. They write everything, but they use AI to bring it to life. And there's parts of their music that, is super powerful. And for folks with CRPS, um, there's sections in one of their songs where they talk about having phantom limb and, and going through the pain and not being really heard.
Speaker 3: And that is a connection for people that can be life saving, right? If you can connect with somebody on that level, it can really be that powerful. So it's not just a game or an opinion. There's life behind it.
Speaker 1: Right, right.
Speaker 2: Very well said. Absolutely.
Speaker 3: All right.
Speaker 2: Well, thank you, Jenny.
Speaker 3: That's my two cents. I'll see you later.
Speaker 2: Okay. Bye-bye. All right. Very good.
Speaker 1: Jenny leaves us. And we will open the phone line up. We only have really a little bit of time left in this hour. But if you do happen to be listening live on Saturday morning and you would like to chime in on the subject, 603-250-6007 is the number.
Speaker 3: 603-250-6007.
Speaker 1: Like I said, we don't have a lot of time, but I know a lot of musicians listen to the show. So if you happen to be a musician listening and you have a thought on the subject, 603-250-6007. It feels like old times. I don't give the number out much since we moved to Saturday mornings, because the show's a little different than it used to be. But when we were on in the afternoons, we would take a lot of calls. Some of them would be good. Some of them would be not, uh, anyway, but we won't, we won't get into all that.
Speaker 2: Um, so, uh, just getting back to the article just to finish up on this.
Speaker 1: So like I said, uh, you know, Deezer, uh, back in, um, 2025, they found that 85% of the live streams, uh, generated by fully AI generated were fully AI generated tracks. Um, uh, Again, this is from musicindustryworldwide.com. Deezer's disclosure in April included that nearly 44% of its daily music uploads were AI-generated. Deezer said in July that that share had passed half in June, so now more than 50%, with close to 90,000 fully AI-generated tracks arriving a day. That's staggering, too, when you think about it.
Speaker 2: 90,000 full.
Speaker 1: That means 90,000... fully AI-generated tracks being uploaded to Deezer every single day. Fitzgerald asked the office to, quote, examine the prevalence of streaming fraud across digital platforms and how it may be affecting the music industry as a whole, and specifically the effects of AI-generated streaming fraud on the licensing and distribution of sound recording, performance, and mechanical royalties, unquote. The notice was published one day after Michael Smith, now we talked about this on the show too, you might remember this story, One day after Michael Smith was sentenced to 18 months in prison for a scheme...
Speaker 1: Now, this was definitely fraud. For a scheme that used bot accounts to stream hundreds of thousands of AI-generated songs. So what this guy did was, Michael Smith, from what I can recall from the story, because we did cover this on the show, he... created he developed a whole system where he created all these different accounts with all these different ai completely ai generated artists with all ai generated music uploaded all of it and then created all these bots to go and stream all of this music And that generated royalties, of course, which amounted to millions.
Speaker 1: I mean, this guy, this whole scheme, this guy generated millions of dollars by uploading AI, completely AI-generated music, not real artists, and then creating bots to go and listen to that music. And it's impressive when you think about the scale on which he had to do all of that to make it work and actually generate millions of dollars. I mean, that's pretty impressive. I'm not going to get, it does get into a little bit of the details.
Speaker 2: Okay.
Speaker 1: Smith of Cornelius, North Carolina pled guilty in March to one count of conspiracy to commit wire fraud. And it was conspiracy, by the way. Apparently he had a couple of other people involved.
Speaker 2: In this with him.
Speaker 1: But I mean, it's, it's just amazing that anyone could even get away with that. And this was over the course of several years, uh, you know, to generate millions of dollars in royalties from these, um, AI generated tracks that were not being listened to by actual people. Uh, on October 6th, US judge, uh, John G. Co, uh, Coelt, Colt, uh, ordered Smith to forfeit, uh, over $ 8 million. The last figure agreed by both sides. That's how much this guy generated in fraudulent, uh, royalties. Uh, the notice states that the primary harm of streaming fraud is the dilution of the royalty revenue pool.
Speaker 1: Um, the US CEO also points to secondary harms, including effects on playlist placement and the streaming data that artists and venues rely on when planning tours.
Speaker 2: That's a very good point.
Speaker 1: Among the notices, 10 subject, actually we should stop on that for a moment because if you are in that part of the music industry where maybe you're a promoter and I used to do a lot of that, a lot of promoting events, uh, Kind of a nightmare in a lot of ways. We'll probably stop on this point because we've got to get to a break before the second hour. But if you were a promoter and you're trying to decide on who to bring to your venues, one of your venues for a show, and you're trying to decide, you know, when you're a promoter, you take a financial risk.
Speaker 1: You book the event, you hope it makes enough money, and you hope you make a profit.
Speaker 2: So if one of your data points....
Speaker 1: When deciding on whether or not to book an artist is how many streams they have. And if those streaming numbers aren't necessarily legitimate because.
Speaker 2: It's all bots....
Speaker 1: Then that's a problem that skews you know you're you're working with data that's not accurate in other words it's no different than um you know say you're an advertiser and you're looking um you're looking at a potential YouTube streams of a podcast that you might sponsor and, and, uh, but their numbers are fake because they're actually using bots to inflate artificially inflate their numbers. You know, then the product that the, uh, or I should say the, the podcast that you're potentially going to be advertising on, uh, And I use this as an example because it happens all the time.
Speaker 2: All the time.
Speaker 1: The podcast that you're considering advertising on might not really have those listeners or viewers. Those numbers are artificially inflated by buying bots.
Speaker 2: There's all kinds.
Speaker 1: Of ways this goes on. This is a little bit of a side street, but people use that for their businesses. They buy Google reviews. You can actually hire a company that will generate fake Google reviews, positive Google reviews for you. Um, now if you get caught and people tend to get caught because Google's, uh, detection methods are sophisticated enough, they can actually tell when, when these reviews are fake, they have ways, they have ways, uh, you know, then they'll deactivate your Google ads account, which no business wants that because that sucks.
Speaker 2: But anyway, this kind of thing goes on all the time.
Speaker 1: And how it affects the music industry and how prevalent it is in the music industry is certainly a problem. But we don't have time for the rest of this. We're going to have to get to a break. There's a lot here. But if you want to look at the entire article, again, it's on musicbusinessworldwide.com, titled Music Streaming Fraud Inquiry Launched by U.S. Copyright Office. And that will give you all the dirty details.
Podbean