AI country “artists” Cain Walker and Breaking Rust
This will not solve the scourge of AI music infiltrating the marketplace, or slow the increasing flood of it, or be a fool proof tool that still can’t be manipulated. But it’s a very important start to addressing a very big problem that will help protect creators and inform consumers.
Announced last week, a host of major organizations have presented a plan to mark AI-generated and AI-assisted tracks across all platforms and mediums so that consumers know when their listening to something utilizing the emerging tech.
The IFPI (International Federation of the Phonographic Industry), RIAA (Recording Industry Association of America), The Recording Academy (Grammys), A2IM (The American Association of Independent Music), WIN, IMPALA, SAG-AFTRA, and the Human Artistry Campaign all forwarded a universal labeling system they hope the entire music industry adopts globally.
AI-Generated (track-level label) Generative AI was used to generate the entirety or the primary portion of the creative elements of the recording. This would include, for example:Lead vocal performance generated by AI, or Key instrumental performance generated by AI, or Entirely prompt-generated AI music.
AI-Assisted (track-level label) The recording was created substantially by humans and expresses human creativity; however, generative AI was used for some expressive elements. Humans performed the lead vocal and primary instruments.
As Saving Country Music implored back in November of 2025, “No different than when a track is uploaded to streaming services and it’s mandated to disclose whether explicit language is present in it, the same should go for disclosing AI, with this information then presented to consumers … Your favorite human musicians who are in an existential battle with AI-generated slop, both for positions on the charts, and for attention on social media. If we did something to address this issue right now, we would already be months, if not years behind.”
Nine months later, something is finally being done, and it couldn’t be soon enough. According to streaming service Deezer, 44% of the music delivered to their platform is AI-generated. Deezer is one of the first companies that implemented software to help detect AI-generated tracks. Apple Music also says that over one-third of the music uploaded to its platform is 100% AI.
Back in November, the concern was that by Q2 of 2026, there might not be any human-based music industry left. At the rate things were progressing, that’s where we were headed. But what was not considered was the dramatic public backlash against the technology, not just in music, but across the AI spectrum, from opposition to data center construction on the local level, to distaste for “AI slop” videos mucking up social media.
Public rebuke of AI has been the saving grace for human musicians. That is why implementing a tool to identify AI-generated tracks is so powerful.
“As AI continues to be integrated into the creative process, artists and fans alike deserve a clear way to communicate how and when it’s being used,” says Harvey Mason Jr., The Grammys CEO. “This initiative ensures that creativity, authorship, and artistic intent remain at the center of every song. Giving artists the ability to tell that story strengthens trust and supports a more sustainable future for music.”
Something this tool also protects is human creators getting accused of being AI—an accusation that can have damaging effects. There are already multiple companies offering certifications for human-made songs to protect creators from these kinds of accusations, but the kind of universally-adopted system proposed would have more power than a patchwork of independent entities.
If this system is implemented industry wide, will it be foolproof? Of course not. Will AI creators try to game it? Absolutely. From now on, it will be a cat and mouse game between streaming services, distribution platforms, and awards/charting organizations with software looking to detect AI and the AI creators trying to circumvent it. That a guarantee.
But you have to start somewhere, and an honest attempt to inform the public so they can make better choices is an excellent place to start.
The public should be free to choose what they listen to, even if it’s AI-generated, which I do not listen to by the way.
However, proper labelling is a great step, and may be the only step, that can be taken to at least inform said public. Just like food labeling, read it and decide if you eat it.
Now the question: If a real artist uses AI “in part” in their music, should this be noted? A tangled web indeed.
What I love about this idea is it draws a distinction between a fake artist with a totally 100% AI song, and someone who might use AI as a glorified thesaurus or research tool, or use it in the demo process, etc., or even uses it to complete a line maybe they’re stumped by in a lyric. When I wrote the AI policy for Saving Country Music, I tried to account for this, because it really is two separate things. If it was up to me, there would be zero AI. But just like anything, there’s a spectrum you have to account for.
Hey I loved the article. Personally I have no problem with AI if it’s labeled. Personally I have no desire to listen to something not created by humans. I just hope artist still make real art and not take shortcuts
Oh and Trigger one thing, I love what you do and I’m not trying to be a grammar Nazi but I think you made an error.
“Apple Music also says that over one-tired of the music uploaded to its platform is 100% AI.”
Over 1/3rd!!! That is bonkers. I had no idea how prevalent it is. I imagine that a lot of that is music that is more digitally influenced, but still that is crazy.
I had a patient that played me her new song on her phone and it was incredibly impressive until she told me that the vocals and rhythm section was AI. I was only then mildly impressed that she could manipulate the software that well.
In country, I imagine that most all of us want to hear the real thing – and that includes arrangements.
I’m surprised it isn’t more than 1/3rd honestly. You have content farms that are just posting AI slop on YouTube trying to game the system in an attempt to get enough views on their slop to make a buck with no work involved.
My guess is the reason it isn’t higher is either:
1. The increased cost of AI “tokens” from the various models eliminates some of the profit motive. Either to push out slop if AI companies are heavily subsidizing your usage so that the “math, math’s”.
2. Not enough money to be made pushing AI slop onto streaming platforms due to the low play rates and a scammer can make more money creating AI slop that defrauds grannies on Facebook than getting spins for AI songs.
Trig, what about the very realistic possibility that real human-created music will be removed from the platform when accused of undisclosed AI involvement?
Good point. I absolutely think this could be a concern, and as I said in the article, I am already seeing artists not using AI getting attacked for doing so based off assumptions, which can be very damaging to their reputation. Though I think we should draw a hard line against AI, I think we need to be very careful with these accusations.
One thing though, right now, none of these platforms are removing AI at all. So we’ll see.
The reality is you will always have “false positives” when do are moderating anything. This is no different to me than moderation of social media platforms. Back when places like Twitter (X) or Facebook did moderation or even YouTube to this day, you had posts or videos pulled down due to wonky scanning technology or flat out human error. Not because the content violated platform standards.
You will have this will AI scanning as well. At the end of the day though, most people *dont* want to listen to AI slop music. Same way *most* people want a level of moderation in their social media feeds. If they didn’t, X would be gaining users at a rapid rate and 4chan/8chan would be bigger than Reddit.
The key is having a level of transparency from the streaming platforms on how this is decided and clear steps for folks to appeal if their is an error. I’m not holding my breath on this as Spotify, Amazon, Apple, YouTube, etc. are not exactly famous for their level of transparency and willingness to say “my bad” when they get it wrong. So sadly I do think some legit albums/artists/songs will get caught up in this, such is the nature of these things.
That’s a crazy stat that a third of music being uploaded to music streaming platforms is already AI slop.
This new labeling protocol is a good start for the music industry. Meanwhile, I wish social media would introduce some kind of opt in or opt out setting for AI-generated socials posts. Nowadays, the majority of posts in my Facebook feed, in particular, are clearly AI-generated. And much of the rest of the feed is ads, so there’s not a lot of reason for, or enjoyment in, scrolling that feed anymore.
I am so tired of the weird AI duets. It’s always Jelly Roll and So & So, singing weird Christian songs. I politely pointed it out to a ppl sharing this trash, THEY DON’T CARE. It sounds good .CTFO I want to gag! Oh, & I nearly forgot the recent Ella Langley stuff, the picture doesn’t even look like her. i.can’t.people.no.more.
July 13, 2026 @ 8:42 am
The public should be free to choose what they listen to, even if it’s AI-generated, which I do not listen to by the way.
However, proper labelling is a great step, and may be the only step, that can be taken to at least inform said public. Just like food labeling, read it and decide if you eat it.
Now the question: If a real artist uses AI “in part” in their music, should this be noted? A tangled web indeed.
July 13, 2026 @ 10:42 am
What I love about this idea is it draws a distinction between a fake artist with a totally 100% AI song, and someone who might use AI as a glorified thesaurus or research tool, or use it in the demo process, etc., or even uses it to complete a line maybe they’re stumped by in a lyric. When I wrote the AI policy for Saving Country Music, I tried to account for this, because it really is two separate things. If it was up to me, there would be zero AI. But just like anything, there’s a spectrum you have to account for.
July 13, 2026 @ 8:46 am
It’s a start. It has to be better than nothing. I have my doubts as to whether it will be effective but I do hope it works
July 13, 2026 @ 9:03 am
I don’t see anything addressing AI generated lyrics or melodies. It only seems to address performance.
July 13, 2026 @ 10:43 am
I think the second distinction is exactly for that, though maybe they could add to their verbiage.
July 13, 2026 @ 9:11 am
Hey I loved the article. Personally I have no problem with AI if it’s labeled. Personally I have no desire to listen to something not created by humans. I just hope artist still make real art and not take shortcuts
Oh and Trigger one thing, I love what you do and I’m not trying to be a grammar Nazi but I think you made an error.
“Apple Music also says that over one-tired of the music uploaded to its platform is 100% AI.”
July 13, 2026 @ 9:43 am
Good news!
July 13, 2026 @ 9:43 am
Over 1/3rd!!! That is bonkers. I had no idea how prevalent it is. I imagine that a lot of that is music that is more digitally influenced, but still that is crazy.
I had a patient that played me her new song on her phone and it was incredibly impressive until she told me that the vocals and rhythm section was AI. I was only then mildly impressed that she could manipulate the software that well.
In country, I imagine that most all of us want to hear the real thing – and that includes arrangements.
July 13, 2026 @ 10:58 am
I’m surprised it isn’t more than 1/3rd honestly. You have content farms that are just posting AI slop on YouTube trying to game the system in an attempt to get enough views on their slop to make a buck with no work involved.
My guess is the reason it isn’t higher is either:
1. The increased cost of AI “tokens” from the various models eliminates some of the profit motive. Either to push out slop if AI companies are heavily subsidizing your usage so that the “math, math’s”.
2. Not enough money to be made pushing AI slop onto streaming platforms due to the low play rates and a scammer can make more money creating AI slop that defrauds grannies on Facebook than getting spins for AI songs.
July 13, 2026 @ 9:49 am
Trig, what about the very realistic possibility that real human-created music will be removed from the platform when accused of undisclosed AI involvement?
July 13, 2026 @ 10:45 am
Good point. I absolutely think this could be a concern, and as I said in the article, I am already seeing artists not using AI getting attacked for doing so based off assumptions, which can be very damaging to their reputation. Though I think we should draw a hard line against AI, I think we need to be very careful with these accusations.
One thing though, right now, none of these platforms are removing AI at all. So we’ll see.
July 13, 2026 @ 10:53 am
The reality is you will always have “false positives” when do are moderating anything. This is no different to me than moderation of social media platforms. Back when places like Twitter (X) or Facebook did moderation or even YouTube to this day, you had posts or videos pulled down due to wonky scanning technology or flat out human error. Not because the content violated platform standards.
You will have this will AI scanning as well. At the end of the day though, most people *dont* want to listen to AI slop music. Same way *most* people want a level of moderation in their social media feeds. If they didn’t, X would be gaining users at a rapid rate and 4chan/8chan would be bigger than Reddit.
The key is having a level of transparency from the streaming platforms on how this is decided and clear steps for folks to appeal if their is an error. I’m not holding my breath on this as Spotify, Amazon, Apple, YouTube, etc. are not exactly famous for their level of transparency and willingness to say “my bad” when they get it wrong. So sadly I do think some legit albums/artists/songs will get caught up in this, such is the nature of these things.
July 13, 2026 @ 10:00 am
That’s a crazy stat that a third of music being uploaded to music streaming platforms is already AI slop.
This new labeling protocol is a good start for the music industry. Meanwhile, I wish social media would introduce some kind of opt in or opt out setting for AI-generated socials posts. Nowadays, the majority of posts in my Facebook feed, in particular, are clearly AI-generated. And much of the rest of the feed is ads, so there’s not a lot of reason for, or enjoyment in, scrolling that feed anymore.
July 13, 2026 @ 12:34 pm
I am so tired of the weird AI duets. It’s always Jelly Roll and So & So, singing weird Christian songs. I politely pointed it out to a ppl sharing this trash, THEY DON’T CARE. It sounds good .CTFO I want to gag! Oh, & I nearly forgot the recent Ella Langley stuff, the picture doesn’t even look like her. i.can’t.people.no.more.