AI Music Revolution

The Real AI Music Problem Has Nothing to Do With AI

• Josh • Episode 19

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 15:17

Send us Fan Mail

The AI music panic doesn't match the math. The global music industry generated $105 billion in 2023. AI music accounts for about 1.5% of actual streams. For the median independent artist, the competitive pressure from AI works out to roughly $45 a year. The existential threat framing is wrong — and the creators who understand that have a significant advantage over the ones who don't.

In this episode:

The manifesto — what the actual numbers say about AI and the music industry, why $101 billion of it doesn't care what tool you used, and how to make decisions from data instead of headlines.

The system — most AI music creators are leaving real royalties on the table right now, and it has nothing to do with AI. Four separate royalty streams. Most creators only set up one. Here's what the other three are and how to claim them.

Plus a clip from my conversation with Roy Brennan — a Red Lab Access member from the UK who described Suno as crack cocaine and meant it as a compliment. Roy talks about the moment Suno stopped being a toy and started being a tool, and the single thing that changed everything for him.

The full Roy Brennan interview is available now as Episode 3 of Red Lab Conversations. Link in the show notes.

Red Lab Access — the complete system for serious AI music creators. Books, research, Blueprints, community, and everything we add in the future. One price, lifetime access.

👉 jgbeatslab.com/red-lab-access

Red Lab Conversations drops every Tuesday when we have a story worth sharing. AI Music Revolution drops every Friday.

The Unlock System is JG BeatsLab's methodology for serious musicians working with AI tools. Lane 2 work: human-authored, AI-assisted music creation.

Visit JG BeatsLab: https://www.jgbeatslab.com

The JG BeatsLab Newsletter. One email a week, Thursdays. Studio work, Red Lab research, and methodology in real time. Free signup on the homepage at jgbeatslab.com.

Subscribe to the Newsletter: https://www.jgbeatslab.com/newsletter

Red Lab Conversations is produced by JG BeatsLab LLC, an AI music education company building the methodology, research, and community for serious creators working in Lane 2.

Get more from JG BeatsLab LLC:

Connect:

Contact: josh@jgbeatslab.com

Stop gambling. Start directing.

SPEAKER_01

Hello and welcome to the AI Music Revolution. I am your host, Josh Gillilan, the founder of JG Beats Lab. Here is a number that almost never shows up in the AI music debate. $105 billion. That's what the global music industry generated in 2023. Not streaming revenue, not recorded music, the whole thing. And yet every conversation about AI music acts like the entire industry is under existential threat. Today we talk about what the actual numbers say, why most AI music creators are leaving real money on the table through a completely different problem, and we'll hear from Roy Brennan, a Red Lab Access member from the UK, who describes Suno as crack coquine and meant it as a compliment. This is the AI music revolution. Let's get into it. First up, let's talk about the $105 billion that nobody is doing the math on, because the numbers tell a very different story than the headlines. The music industry had generated $105 billion in 2023, the last year we have data for. That's the whole ecosystem: live events, sync licensing, merchandise, publishing, recorded music, performance rates, everything. Of that, $105 billion, streaming accounts for roughly $20 billion. That's the segment everyone is fighting about. That's the revenue pool that AI music is supposedly threatening. So before we go further, we're already talking about $20 billion out of a $105 billion industry. So about 19% of the total industry. Now let's talk about what AI music actually represents inside of that $20 billion. Deezer reported that AI generated music accounts for about 18% of uploads to their platform. That number gets cited constantly as evidence that AI is taking over music. Here's what that number doesn't tell you. Uploads are not streams. Actual AI music listening share on major platforms sits around one and a half percent of total streams. 18% of uploads, one and a half percent of listen. That gap between those two numbers is the gap between a flood of content and an audience that has actually chosen to engage with it. Apply that 1.5% to the global streaming revenue pool, and you get roughly $306 million currently flowing to AI-generated music across all platforms. That's the actual size of the competitive threat in dollar terms today. $306 million in a $105 billion industry. And for the median independent artist, someone earning around $3,000 a year from streaming, the portion theoretically displaced by AI competition works out to about $45 per year. The existential threat framing doesn't match the math. And here's the part that really matters here. $101 billion of this industry doesn't care whether your track was made with a guitar or a generative model. Live music doesn't care. Sync licensing doesn't care. It cares whether the track fits the brief. Publishing revenue doesn't care. The entire argument against AI music is concentrated in the smallest, most commoditized segment of the industry. The per stream royalty on a platform that pays fractions of a cent per play. Know the actual numbers. Make your decisions from there. Because the creators who are going to win the next decade aren't the ones who successfully lobbied to slow down AI adoption. They're the ones who understood the landscape clearly and built catalogs and revenue streams that aren't entirely dependent on Spotify pennies. Next up, speaking of revenue streams, most AI music creators are leaving real money on the table right now. And it has nothing to do with AI or streaming share. It's a registration problem. And it's completely fixable. Let's get into it. Here's the thing that nobody tells you when you first upload a track to DistroKid. That upload covers one of four royalty streams, just one. DistroKid is your distributor. It gets your music onto Spotify and Apple Music and about a hundred other platforms. And it collects your master recording royalties from those platforms. That is genuinely important. But the other three streams, those require completely separate registration that DistroKid does not handle for you. Your pro. This is BMI or AskAut if you're in the United States. They collect songwriting performance royalties when your music is played publicly on streaming, on radio, in venues, on TV. If you haven't joined a pro and registered your songs, that royalty stream isn't being collected. Nobody is holding it for you. Next, you have the Mechanical Licensing Collective, or the MLC. They collect digital mechanical royalties from interactive streaming services. This is separate from what your pro collects. Registration is free. Most independent creators never register, which means those royalties sit there unclaimed. And in some cases, you cannot go back and reclaim them retroactively once the window closes. Finally, you have sound exchange. They collect digital performance royalties on the master recording from non-interactive services. Think satellite radio or internet radio or webcasters. Not the same as Spotify royalties, a completely separate system. Registration is free. Most creators skip it entirely. Four registrations, four separate royalty streams. If you're only set up with DistroKid, you have one of the four. The other three are sitting there waiting for you to claim them. I wrote all this up in the new Unlock Music Rights and Registration book. 90 plus pages covering all four systems, and what each one collects, how to register, and how to make sure AI assisted music qualifies. $12.99 at jgbeatslab.com. And if you want personal guidance as you run your first release, the release support package adds 60 days of direct email support from me as you work through each step. Link is in the show notes. Finally, let's hear from Roy Brennan. Roy is a Red Lab Access member from the UK who has built one of the most sophisticated AI music production ecosystems I've seen anywhere in this space. Connecting Claude projects to Notebook LM via MCP, building musicology protocols that translate producer aesthetics like Tony Visconti into Suno tag language, stripping novels for narrative scenes before a single note is even generated. But before all of that, there was a simpler moment. The moment Suno stopped being a toy and started becoming a tool. Here's Roy. What was your entry point or exposure to Suno? What did that look like for you?

SPEAKER_00

I was um exposed to sort of the chatbots first and um spent a short while um building what I'd call um augmented role um projects in in uh cloud and chat GPT. That is to say, how can uh how can a machine or an AI tool not replace you but create like an augmented version, a more organized version, a more rounded thinking uh to the way that you uh approach your role. So it's to augment the efficiency of a job rule. I discovered soon about just because somebody said you've got to hear this, it was one of those, it's while what the type of prompting and the sun comes out kind of I I wasn't sort of so taken with it at first that that I jumped straight in and and and subscribe or anything like that, but it wasn't long before I did try it for a bit, and it's it's it's like crack cocaine. We everybody knows that, right? But it's follow without any sense of purpose or meaningful direction and structure. There's only so much, so many tunes that you can actually do and sit there waiting to think does somebody like it or not like it. Um that this is the key thing with Vader. It was a little bit like a sidewind and missile that that woke me up to something I kind of must have known at some point, but and that's the structure. Um a song is meaningless if it's got no purpose and intent. It it's it you need and have to have something to say. You know, the moment I used Vader, that was the moment that I just got obsessed with Suno, but only within the context of doing it the way that I've decided to pursue it.

SPEAKER_01

That was Roy Brennan. And I want to sit with what just happened across this entire episode for a second. We started with the numbers. The AI music panic doesn't match the math. $101 billion of this industry doesn't care whether your track was made with a guitar or a generative model. The existential threat framing is wrong. The industry is fine. Then we talked about the money most creators are leaving on the table right now. Not because of AI and not because of streaming, but because they uploaded to DistroKid, thought they were done, and walked away from the other three royalty streams that nobody told them about. And then Roy showed us what happens when someone actually does all of this right. When you stop generating randomly, build a system with real intention, and wrap the whole thing in a philosophy that keeps you focused and disciplined. That's when the music gets better. That's when Happy Roy shows up. The industry isn't dying. The royalties are waiting, and the machine will do exactly what you tell it to do. If you want to hear the full Roy Brennan conversation, the notebook LM workflow, the MCP connection, the Tony Visconti protocol, all of it, that's episode three of the Red Lab Conversations. It's out now. Links in the show notes. If today landed for you, if you're starting to think differently about what it actually means to build something real in this space, Red Lab Access is where that system lives in full. The books, the research reports, the blueprints, the community of creators who are doing this work every single day. One price, lifetime access, everything included now, and everything we add going forward. jgbeatslab.com slash red access. Links in the show notes. Red Lab Conversations drop every Tuesday when we have a story worth telling. AI Music Revolution drops every Friday. Subscribe so you don't miss either one. The industry isn't dying. The royalties are waiting. And the machine will do exactly what you tell it to. See you next time.