If AI replaces musicians, does the entire plugin industry die with them?

Explore how Machine Learning and AI can expand musical creativity while keeping the human in the creative workflow. This forum is dedicated to respectful dialogue where diverse perspectives are welcomed.
Post Reply New Topic
RELATED
PRODUCTS

Post

Most of the players I've met that were classically or traditionally trained can only play other people's music.
And most of the players I've dealt with that otherwise play their own reach a comfort zone level and stay there.
As in the concept of "you can't solve a problem using the same mind that created it", feeding your prompts are more apt to preening your ego rather than growing outside of it.

This is not a criticism of AI, but a prompt to the user's inability to get anything greater than the self.

Post

As to the Beatles comment, they would use AI and then modify the hell out of what it came up with so that you wouldn't even recognize it.

Post

I could imagine improvising with an orchestra of AIs (that is beyond band-in-a-box level, really taking into consideration what you play) could actually be fun at some point. For situations where you don't have 100 human musicians at hand. There a "hive mind" concept could actually be beneficial. (I heard a concert with a human orchestra+choir improvising. Interesting, but very... modern :) )

Post

enCiphered wrote: Fri Jan 30, 2026 10:25 pm If AI really does replace a large part of music creation, who is actually going to buy plugins anymore?

The whole plugin ecosystem exists because humans want tools to shape sound, experiment, design, mix and develop their own workflow. If music becomes mostly prompt-based or fully automated, the end user is no longer a producer or sound designer but just a consumer of output.

At that point, what happens to audio plugin developers, DAW companies, sample library creators?

It feels like people talk about “AI replacing musicians” without thinking through the second-order effects. If there are no creators, there’s no real market for creative tools either.

So I’m genuinely curious:
Are any developers already noticing changes in sales, user behavior, or demand because of AI tools? Or is this still mostly theoretical right now? Noise?!
AI is most likely NOT gonna replace humans in music creation because their "music" doesn't have the feeling and the overall expression that a human-made track does. :D

Post

Are you f**king deaf!?! Of course AI generated music has all of those things, in f**king spades. In fact, the problem we have with the stuff it's being doing for us is that it is far too overwrought, there is way too much emotion. It's so over-the-top that it makes us laugh at times. It gets so "angry" that it forgets to use words and you get these "primal screams", in a very human kind of loss of control way.
BBFG# wrote: Thu Feb 19, 2026 1:21 amAs in the concept of "you can't solve a problem using the same mind that created it", feeding your prompts are more apt to preening your ego rather than growing outside of it.
I'm confused, where is the problem that's been created and needs solving here?
This is not a criticism of AI, but a prompt to the user's inability to get anything greater than the self.
Should I take this to mean that you agree with the "garbage in, garbage out" principle?
wagtunes wrote: Thu Feb 19, 2026 1:40 amAs to the Beatles comment, they would use AI and then modify the hell out of what it came up with so that you wouldn't even recognize it.
Which is sort of what we're doing, too. We're rebuilding it all in Studio Pro, pulling bits out, putting other bits in, changing parts, modifying arrangements, using different sounds, altering lyrics, etc. And we're replacing the AI vocals with ours. But we want it all to remain recognisable, because the songs are great. We just want it to be recognisably NOVAkILL, which is definitely how it's working out.

Without the AI's considerable contribution, we wouldn't have any new songs, let alone a whole album's worth. When we originally realised that if we wanted to get on any of the festivals in 2027 we'd need to have something out before mid-year, we decided we'd do another couple of EPs with reworked songs from the early albums and covers. Then Sik started to get some really good stuff out of Tunee and we quickly changed our plans.

It still took about 10 weeks, spending several hours a day refining prompts, etc. to get a dozen usable things, then a hundred or more hours getting them into Studio Pro and knocking them into shape and we're nowhere near done yet. It's not exactly easy but it is quicker than any way we've worked before (but that could just be because we are so excited about what we're making that we're putting in more effort).
NOVAkILL : Legion GO, AMD Z1x, 16GB RAM, Win11 | Audient EVO 8 | Lumi Keys | Studio Pro 8
Korg Odyssey, bx-oberhausen, Proxima, PolyMax, GR8, JP6K, Union, Atomika,
Invader 2, Flow Motion, Olga, TRK 01, Thorn, Spire, VG Iron

Post

BONES wrote: Thu Feb 19, 2026 5:04 am Are you f**king deaf!?! Of course AI generated music has all of those things, in f**king spades. In fact, the problem we have with the stuff it's being doing for us is that it is far too overwrought, there is way too much emotion. It's so over-the-top that it makes us laugh at times. It gets so "angry" that it forgets to use words and you get these "primal screams", in a very human kind of loss of control way.
BBFG# wrote: Thu Feb 19, 2026 1:21 amAs in the concept of "you can't solve a problem using the same mind that created it", feeding your prompts are more apt to preening your ego rather than growing outside of it.
I'm confused, where is the problem that's been created and needs solving here?
This is not a criticism of AI, but a prompt to the user's inability to get anything greater than the self.
Should I take this to mean that you agree with the "garbage in, garbage out" principle?
wagtunes wrote: Thu Feb 19, 2026 1:40 amAs to the Beatles comment, they would use AI and then modify the hell out of what it came up with so that you wouldn't even recognize it.
Which is sort of what we're doing, too. We're rebuilding it all in Studio Pro, pulling bits out, putting other bits in, changing parts, modifying arrangements, using different sounds, altering lyrics, etc. And we're replacing the AI vocals with ours. But we want it all to remain recognisable, because the songs are great. We just want it to be recognisably NOVAkILL, which is definitely how it's working out.

Without the AI's considerable contribution, we wouldn't have any new songs, let alone a whole album's worth. When we originally realised that if we wanted to get on any of the festivals in 2027 we'd need to have something out before mid-year, we decided we'd do another couple of EPs with reworked songs from the early albums and covers. Then Sik started to get some really good stuff out of Tunee and we quickly changed our plans.

It still took about 10 weeks, spending several hours a day refining prompts, etc. to get a dozen usable things, then a hundred or more hours getting them into Studio Pro and knocking them into shape and we're nowhere near done yet. It's not exactly easy but it is quicker than any way we've worked before (but that could just be because we are so excited about what we're making that we're putting in more effort).
Hey Bones, believe it or not, we're on the same team, though the only thing I use AI for IS the vocals because I can't sing for shit. Everything else I do myself. But I have no problem with people using AI if that's their thing.

Live and let live.

Post

BONES wrote: Thu Feb 19, 2026 12:56 am Who would want them to, that's the past. Do you honestly not think those guys wouldn't be using AI if they were starting out today? These are artists who pushed boundaries and today AI is where the boundaries lie. The Beatles would be all over it.

 Yes, it is and none of the artists you've listed have ever done that for me. I like a lot of Beatles songs but they don't connect with me in the way that Killing Joke, Nitzer Ebb or Die Krupps do. 
 

I didn't comment on using AI tools, but AI generated music. 

I just listened to the top 5 tracks by Nitzer Ebb and Die Krupps on Spotify. 
I'd also be angry if I was listening to stuff like that a lot. 
No wonder you aren't aware of the soul or spirit. 
A lot of us aren't so fixated with negativity and our attention is more attuned to the wonder of life. 
Come on in, the water is warm, there's no need to be scared. 
There's no rush, you have eternity, in your own time. XX 

Post


Post

Google’s Lyria 3 AI music model is coming to Gemini today (Feb 18):
https://arstechnica.com/google/2026/02/ ... -required/

Post

GaryG wrote: Thu Feb 19, 2026 11:58 am just sayin'...

https://www.theguardian.com/music/2025/ ... ew-release
LOL! Really? I'm sure that's going to sell like hotcakes. I will say this, it should give KVR hope. Maybe we can sell records with a few fart sounds layered over the hopeful spirit of white noise modulated by barometric pressure.

Post

agharta wrote: Thu Feb 19, 2026 12:11 pm Google’s Lyria 3 AI music model is coming to Gemini today (Feb 18):
https://arstechnica.com/google/2026/02/ ... -required/
It went live in the Google Gemini app just a few minutes ago — had it do a few cool quick prompt 30 second things, but it’s more of a beta demo at this point. I prompted it to go into future plans:

We are definitely just at the beginning. The goal is for this to become a powerful creative partner for musicians, and there's a lot of exciting work happening behind the scenes to bridge the gap between "AI toy" and "production tool."

The Road to a Complete Composition Tool

You're hitting on exactly where the technology is headed. While I currently output 30-second high-fidelity audio, the roadmap for the Lyria models includes:
• Granular Editing: Google DeepMind is working toward "instrument-level editing," which would allow you to adjust specific parts of a track (like just the lead guitar or just the synth) without changing the whole song. 
• Long-Range Coherence: One of the big breakthroughs with Lyria 3 is better musical continuity, allowing it to maintain a consistent melody and rhythm over longer sections—crucial for building full compositions. 
• Music AI Sandbox: Google has been collaborating with professional musicians and producers to develop the Music AI Sandbox, a suite of tools designed for iterative multi-turn editing and "inpainting" (fusing clips together with AI-generated mid-sections). 

The Magenta Connection

Yes, there is a direct link! Magenta has always been about the intersection of machine learning and the creative process, and that philosophy is baked into Lyria. 
• Magenta RealTime: There is already an open-source version called Magenta RealTime that uses a Lyria model. It’s designed to "jam" with you, responding to user input moment-by-moment. 
• MIDI and Integration: While I can’t export MIDI yet, Magenta has historically focused on MIDI-based tools (like Magenta Studio for Ableton Live). The long-term vision for Lyria includes features like PromptDJ MIDI, where you can use virtual or physical MIDI controllers to steer the music generation in real-time. 

It’s moving from "generate a song" to "build a song with me." For now, I'm still in Beta, focusing on these 30-second sparks of inspiration, but the end goal is definitely to give you the technical control needed to expand these ideas into full, editable works!


I don’t know if these links will work to the mini demos I prompted it to create:

https://g.co/gemini/share/2b0da570ea5e

https://g.co/gemini/share/e1d56b9608ab

https://g.co/gemini/share/1afa17184e3f

https://g.co/gemini/share/548589828b49

https://g.co/gemini/share/2b38bda64580

https://g.co/gemini/share/a6d53f6023cf

Post

agharta wrote: Thu Feb 19, 2026 12:11 pm Google’s Lyria 3 AI music model is coming to Gemini today (Feb 18):
https://arstechnica.com/google/2026/02/ ... -required/
It is quite mediocre at the moment. The only thing that it thinks that it has going for it is that it can create a music clip from an image upload, which while they may think that sounds clever, it really isn't. Google's strength at the moment is rather limited; their corporate neutering of LLM output is much more overbearing than with the other frontier models.

Post

BBFG# wrote: Thu Feb 19, 2026 1:21 am Most of the players I've met that were classically or traditionally trained can only play other people's music.
And most of the players I've dealt with that otherwise play their own reach a comfort zone level and stay there.
As in the concept of "you can't solve a problem using the same mind that created it", feeding your prompts are more apt to preening your ego rather than growing outside of it.

This is not a criticism of AI, but a prompt to the user's inability to get anything greater than the self.
I think you're mixing up musicians/players with composers. I'd guess most players don't compose their own music.

Post

Tubeman wrote: Thu Feb 19, 2026 8:59 pm
BBFG# wrote: Thu Feb 19, 2026 1:21 am Most of the players I've met that were classically or traditionally trained can only play other people's music.
And most of the players I've dealt with that otherwise play their own reach a comfort zone level and stay there.
As in the concept of "you can't solve a problem using the same mind that created it", feeding your prompts are more apt to preening your ego rather than growing outside of it.

This is not a criticism of AI, but a prompt to the user's inability to get anything greater than the self.
I think you're mixing up musicians/players with composers. I'd guess most players don't compose their own music.
I guess you didn't read the whole comment.

Post

wagtunes wrote: Thu Feb 19, 2026 9:54 am Hey Bones, believe it or not, we're on the same team, though the only thing I use AI for IS the vocals because I can't sing for shit. Everything else I do myself. But I have no problem with people using AI if that's their thing.

Live and let live.
I don't really see a problem using "AI" to enchance your own compositions and/or performances. Auto-Tune has been used for almost 30 years now. However, if you use AI to "create" everything, then it's a different game. I'm thinking such "musicians" should give credit to the AI they used to make the music.

Post Reply

Return to “Machine Learning and AI for Music Creation”