In Recent Conversation Regarding AI in Music, it has Become Necessary to Take a Stance on the Subject
My Personal Experience with Artificial Intelligence
I’ve not used AI in music directly except to play a bit with Suno and find out a little bit about what it’s capable of. I have used it, and will likely continue to use it for album artwork, I’ve used it for learn coding constructs and it’s taught me new ways of seeing data that I hadn’t thought of before. I’ve never kept that a secret though. I’m going to be controversial here… and that’s okay.
AI can be a wonderful tool. It can teach a subject in ways that human’s can’t. It can cover multiple learning types, create better analogies on the fly. It can teach us things on a level that is exponentially greater than a classroom environment, and there is value to that. It does however have the potential to replace human interpretation, intuition, originality, spontaneity, and there is less value to that for me. I create a lot of software to make this site do things that I need it to do. AI doesn’t build it for me… it could, but that would not help me, because I still wouldn’t understand the code that I’m putting on my server. I choose to have it teach me what I’m doing, and I create the software from that point.
This is not unlike asking a peer to help explain something, and I think this is the highest use of the technology that we could hope for. This is the state of the technology that doesn’t replace us, it betters us. Then there is another state of the technology that needs discussing.
As a Society, We No Longer Choose the Hard Way
This is not a Gen-X vs. Millennial vs Gen-Z debate, it is a personal fact… My personal fact, that I’ve observed, for myself. We should have raised our voice harder along side visual artists in the beginning when they complained that their art was being used to train AI models. Yes, that would likely mean a lot of my album art wouldn’t be as quick to generate as it is today. I’m still a graphic design guy, I can draw quite well, and I can still make it happen, but AI is way faster than me with a pencil.
This is one example, aimed at myself of not choosing the hard way. That paragraph is my way of saying, “I get it.” There are only so many hours in a day and they’re all precious. The temptation to use tools like generative AI is real. It comes down to a matter of perspective and, for me, the artwork is far less personal than the audio content. It’s beautiful much of the time, which is good, but it has no emotional bond to me as part of my actual art. Though it is fun to show off.
AI will never be present in my DAW though. That’s a personal choice, because in there, I’m completely myself. That’s where the rubber meets the road for me and we start getting into emotional bonds that matter to me. Not everyone makes music that matters to them, and that’s why there is a use case, I suppose, for AI music.
If you just wanna make a funny song about dancing llamas maybe you don’t need to spend the time learning an instrument. Maybe you don’t need almost $20,000 in gear. Maybe the hard way isn’t for you. Here’s where we get to the controversial part. Your creation is purpose built to make people laugh and while that’s a good thing, it doesn’t need to occupy the same space as Mickey, for example. Mickey practices and hones his craft for hours to be part of the music space, and something auto generated in seconds shouldn’t overtake that. AI has the potential to snuff out and bury real people and I can’t be a part of that.
There’s a Whole Community Out There of People…
I propose this section before I issue a publishing standards update. If you want to make a funny song about dancing llamas, AI is always an option. So is
bigbaldben
Mario Marino
Only the Host
and countless other real people who would generally love to help you. Some of them may charge a fee, but many won’t, because we love the art aspect. I’d totally help too, but I’m generally less funny than the provided options. This is the other downside of AI. It’s socially isolating, if you let it become that way. If you run straight to AI, you miss out on getting to know anyone new. You miss out on getting to celebrate your hilarious llama song with the other people you made friends with during it’s creation. Just food for thought.
Publishing Standards Regarding AI Music
I’m not outright banning it, but I am requiring some conditions. A lot of artists are using AI elements in their music and for those creations where there are real people involved but we have AI sound design elements at play, those elements need to be disclosed to the author that is talking about you on Abrasive Monologues and that author will need to add a disclosure that some elements of your production were created with generative AI.
If your song is completely generated by AI, that will need to be disclosed as well, and the author will issue a disclosure in the article that the entire production is generative AI and that no musicians were involved in the creation.
That way, any potential listeners can decide for themselves whether they want to support your project. While authors are required to disclose AI use in the music they write about, they are also under no obligation to write about it. Each of them is absolutely free to decide for themselves whether spending their time writing about AI music is a worth-while endeavor. That’s a personal choice that should never be made for anyone.
