Creating music with AI - The death of an industry?

AI, Creativity, and the Battle for Originality

I know this newsletter just launched yesterday and is meant to be a weekly edition, but your incredible support has truly inspired me to keep creating. So, here are my thoughts for today—thank you for being part of this journey!

As a songwriter and producer, I’ve always been fascinated by the creative process—its unpredictability, its magic. A song can come to life in five minutes or take years to perfect. But the real magic? Creating something from nothing alongside co-creators. Nothing compares to the feeling of crafting an incredible song with friends. That’s what ignited my love for music as a teenager, and it still drives me today.

However, a few months ago, I actually found myself stepping away from music, prompted by a realization that AI could create music better, faster, and cheaper than any human. An AI can conjure a catalog of 500 beautifully produced songs in just 1-2 days, a task that would take a human a lifetime. I have created about 150 finished songs in my 30 years in the music business. And honestly, I started thinking—what's the point of making any more music?

I see a steep decline ahead for the music industry. The cost of music production is set to drop dramatically, and in just a few years, I can easily imagine artists cloning their voices, uploading them to an AI music generator, and creating an entire album in minutes. The market—already oversaturated—will be flooded with AI-generated music, voices, and even AI-created artists.

For the sync world (music for film), the future looks even more gloomy. Just yesterday, I received a brief from a music supervisor searching for a track for a global commercial, with a licensing fee of up to $650,000. In theory, I could take that brief, paste it into ChatGPT, have it generate the perfect prompt detailing the song’s specifications, and instantly create 100 tracks tailored to the request. Who would ever know? Of course, I’d never do this—but the fact that it’s possible, undetectable, and could disrupt an entire industry is a major, major issue.

Yet, as I reflected on this, I also began to see a deeper complexity. The power of AI in music creation isn't just about speed and efficiency; it's also about the source of its creativity. These AI music generators don’t create from pure originality—they are trained on vast libraries of existing music, learning from and replicating patterns, styles, and sounds. This raises an important question: if AI-generated music is built upon the work of countless artists, how do we navigate the ethical and legal implications? The line between inspiration and imitation becomes increasingly blurred, placing AI-created music in a legal and creative gray zone where unintended similarities could lead to potential disputes.

As much as I eagerly anticipate the democratization of music creation, we need to find a way to compensate the original artists whose work has trained these AI models. But unfortunately we are not there yet in terms of technical advancements. It′s impossible right now to define where the AI generated songs got their inspiration and "sound" from. (Figuring this out could actually be a great business idea!)

Words from the U.S. Copyright Office

Recently, the U.S. Copyright Office clarified that AI-generated content could be eligible for copyright protection if a human significantly contributes to or edits the material. This means that human creativity still holds its ground in copyright law, even as AI tools become more prevalent. They outlined scenarios where AI-generated material can receive copyright protection:

• When human-authored content is incorporated into the AI output.
• When a human significantly modifies or arranges the AI-generated material. (How can this ever be proven?)
• When the human contribution is sufficiently expressive and creative. (OMG! How can this ever be proven?)

While this offers some reassurance to creators using AI, it overlooks the original creators whose work trained these AI models. They should be compensated first, and then we can work on laws for those creating art through prompting. If this does not happen soon, I doubt there will be any new human art in the coming decades. And that saddens me deeply. Especially because of my kids and the coming generations.

So, as we navigate the evolving landscape of AI, what are your thoughts on balancing innovation, creativity, market value and intellectual property rights? Please feel free to comment!

And next, today′s most important AI news:

Latest AI News

1. ByteDance Unveils 'OmniHuman-1': ByteDance has introduced OmniHuman-1, an AI system capable of generating hyper-realistic deepfake videos from a single photo and audio input. This groundbreaking technology can create videos of any length and style, raising questions about the future of digital content and the challenges of verifying authenticity. Read more.

2. Apple Launches AI-Powered Party Planner: Apple has released 'Invites,' an AI-driven app that integrates multiple Apple services to create custom invitations and manage events. This marks Apple's first standalone AI-powered app, showcasing a shift in their AI strategy. Read more.

3. AI in Early Cancer Detection: Johns Hopkins researchers have developed AbdomenAtlas, an AI-powered dataset of 45,000 3D CT scans to accelerate medical image analysis. This innovation could transform early cancer detection by providing comprehensive training data for AI models. Read more.

Top AI Music Tools for Solopreneurs

Need a quick custom created track for your social media post? A jingle? Look no further.

1. Udio.com: Udio.com is an AI music generator that allows users to create professional-quality music in seconds. It’s perfect for solopreneurs looking to produce custom soundtracks or jingles without the need for a full production team. I have used this. It has an unmatched quality in my opinion. I might drop a tutorial about this in the future. Learn more.

2. Suno.ai: Suno.ai is a cutting-edge AI music tool that enables users to create songs from text descriptions. It’s ideal for solopreneurs who want to transform their ideas into professional music effortlessly. It can also create several songs with the same voice. That leads me to think…It will be so easy to create AI artists that can release singles…DAILY!! Learn more.

3. Riffusion: Riffusion is an innovative AI tool that generates music by visualizing sound as spectrograms. It’s a unique solution for solopreneurs looking to experiment with new and creative ways to produce music.I haven′t tried this but it looks super interesting! Learn more.

Thats it for today! Catch you next time! Stay innovative, stay ahead,

Mikkel Solnado