AI + Music --> What will the future look like?

[Bombadil]

Tell me about it!

Don't trust anyone under 30!

[CrystalWizard]

Tell me about it!

Don't trust anyone under 30!

Tim Leary is spinning around the Bardo after reading that.

[Grizzellda]

Reaper... and MuLab and Linux.

[enroe]

We – you, me, all musicians who truly compose – have explored the vast
creative realm of human inventiveness and accomplished incredible things
there – and continue to do so. It has become an important part of our identity.
And is all of that supposed to end now? Because AI, Suno, and the like can
do all of this at a moment's notice, at a "prompt" – and will do it even more
perfectly in the future?

Should we give up all of that—our own creativity and identity—and shrug
and say, "That's just progress"?

I have already invested a lot of time making music and I still enjoy it, so I have no plans to stop.

Yes, I'm very happy about that. I'm curious about your new songs!

I think the only impact A.I. music has on me is to make me emphasize my individuality and make quirkier choices. There is no point in trying to make a perfect dance track, for example, since A.I. is going to excel at that. I might as well pursue my own weird ideas that an algorithm is unlikely to generate.

Interesting perspective. Perhaps this is also the only way for us "real
musicians" to deal with the AI ​​future.

[enroe]

The AI-driven music generation of the future is more of a "final act" that will only
further reduce its significance to virtually zero, as it will be readily available
anytime, anywhere, and will function as "accessory" or "decoration" for other content.

Hmm, yes, that will probably be the case. But it's also the final nail in the
coffin for any human, original emotional composition. And that includes jazz
music—at least in part—and classical music.

That's some story! What it doesn't do at all is support its contentions in any way.
Humans seem to be much more resilient than you know. TBH music must not mean
all that much to you if you're good with saying shit like significance reduced to zero. Speak for yourself.

If you read through the entire preceding section and the argumentation,
you will see that it is all well-founded and derived.

This thread is about the overall impact AI will have on music. If some people
– including myself – analyze the development of AI without prejudice and try
to extrapolate, and arrive at some unpleasant results, then one can examine
the reasoning behind these results. Or one can simply criticize the messenger
for delivering the message. Just like you're doing.

[Grizzellda]

In the old classic western movie, SHANE...

Well Shane is a gunfighter, he thinks he will change his life.

He tells Marion, who is nervous about guns...

"a gun is just a tool, it depends on how you use it".

Same with all this A.I. stuff, it is becoming increasingly tiresome. It is a modern TOOL.

[jancivil]

The AI-driven music generation of the future is more of a "final act" that will only
further reduce its significance to virtually zero, as it will be readily available
anytime, anywhere, and will function as "accessory" or "decoration" for other content.

That's some story! What it doesn't do at all is support its contentions in any way.

If you read through the entire preceding section and the argumentation,
you will see that it is all well-founded and derived.

First of all you quote me as pertains to a speciic statement; I said "this" story doesn't...
I said "this" meaning literally this post. I'd read most of the thread before I typed that (however I didn't follow links to YouTube). It consists of people's conjecture as to a future amidst "AI". Conjecture and opinion ≠ facts.

As an argument, there is no foundation for 'reduce [music's] significance to virtually zero' therein. I'll explain how that is:
"the significance of music" is quite nebulous. The quote I f**ked up attribution to, that you in fact wrote was spun off of a narrative regarding pop music's importance to youth culture. This account of "significance" is much more narrow than the impossibly broad 'significance of music'. Furthermore the narrative of 'lost importance to youth culture' had that as already having occurred.

We'll be arguing at cross purposes unless I, too, take that definition of significance as 'the' definition. I don't. I didn't experience anything of the sort. To me we're talking sociology anyway.
I wasn't old enough to be a hippie, so I didn't have the experience of Grateful Dead & free concerts in Panhandle Park or Golden Gate park. The entirety of significance of this group to me per se was one experience in the mid-1970s, which wasn't about anything but the music.
Since then I heard Weir with Henry Kaiser because I follow Cuneiform Records which I know of because my friends had a record on it.

So for me to even imagine music/youth culture I'm already outside of the subject of "pop music" and I wasn't a part of that particular culture. Pop music's significance to me as a youth was not a social sort of thing and we'll have to apply the label 'pop music' to things much more broadly than I would to make that even work for me.

Your account of things ended up confidently stating that jazz (among other things) will also be reduced to nothing. You can believe whatever things you believe but your narrative supported by what you've picked out of a similar narrative is a circular argument. You don't, you can't know that. There is no support for that anywhere in the vicinity. You have no idea.

Jazz is not youth culture. People in the jazz life aren't bothered by this. You frankly should take the word jazz out of your mouth. You can't make AI do it. There's no 'there', there.
I don't make music that is in any way threatened by the idle experience of music as background. Cultural significance to the masses has absolutely no bearing on the meaning of music for me. People that already make utterly derivative trash may not be able to compete in the marketplace of that. I should care?

[enroe]

@jancivil:

Hmmm ... are you talking about your personal relationship with music?
Or are you (also) talking about the general significance of music in
society?

This is about the latter. That is, it's about stepping out of one's "personal
bubble" and seeing what is happening "in society as a whole." How will
"the music" develop?

If you had actually read the first thread, you would have noticed that it is
divided into 3 different parts:

1. -- AI tools (not relevant here)
2. -- AI song composition (currently)
3. -- AI song composition (in the future, from 2030 onwards)

This is about (3): What will the future look like in a few years? And
Suno & Co. are just the first imitation, the first flash of AI-generated songs.
I think everyone can understand that the rapid development of AI will have
extreme consequences for our entire lives. This includes the world of work,
leisure, communication, self-image – really everything.

The use and significance of music, in particular, naturally varies for each
group in society – there are niches and niche audiences who are hardly
influenced. And then there's the large mass of people who listen to the
radio daily and watch Hollywood blockbusters. Within this "mass,"
"teenagers" represent a special case: On the one hand, they consume
a lot of music. But their consumption also reveals trends for the future.
They often determine what's hip and "in", and what the future of music
looks like. That's why they have special significance. And that's why
major record labels pay particular attention to this group.

----------------------------------------------------------------------------

My hypothesis here – belonging to chapter (3)! – is now:

AI reduces the commercial and societal significance of music composition
to almost zero. Reason: Please scroll back to the initial post and
subsequent ones.

Of course, there will continue to be niche composers and eccentrics who
compose, play instruments, and even mix songs outside the mainstream.
But they will be even more of a fringe group of almost eccentric people
than they are today.

Just like there are many people today who buy vinyl records and enjoy
the large covers. Just like there are people again today who actually
photograph analogously with film and even develop it themselves.
And the younger generation will view these people as a kind of museum
relic. People from the earliest prehistoric times of human history.

[v4p0r]

LLMs are rehashing old patterns, they have a hard time improvising, and if they do this, it sounds stale, silly and otherwise not cool.

I thought something like that, too. AI companies might hire artists to make AI create not only probabilistic variations of "existing" training data, but new things like animations. They do so already and there is all sorts of cool stuff, but in the future we might see the results in practise, which requires them to build better training sets.

Also I think AI providers might replace the "frequency domain based training" (whatever they do now) in order to get rid of sound artefacts. Like, they might acquire synth code and samples from VST developers? Maybe N.I.? In any case, more "custom made synthesis" would be the requirement for me to use AI. Maybe that's unrealistic, because most people just seem to like the current results.

I don't believe in much of the obsolete-argument. I think that AI had access to vast sets of training data for what they do now. People extrapolate it from there, but "The power comes out of the wall socket" the saying goes.

[jancivil]

and, enroe to tell me the reason I don't find that argument supported is because I didn't read what you wrote is obnoxious. Tthe support for an argument being your argument is called begging the question.

If you think "classical music" or jazz will equally meet their doom, all this shows me is you have no idea what that music is or does. AI has no ideas. It goes by prevalence in what it's told to focus on and gives us a fake of it. Music from genuine ideas isn't subject to that.

The argument is essentially people that consume pop music amount to a group mind and lemmings will follow each other over the cliff. This happened with pop a long damned time ago. Copy after copy after copy, quelle surprise, finally the properly homogenous and bland rules OK.

[v4p0r]

Hm, isn't it also like, when the market for buying digital products gets oversaturated, people may simply start selling how they interact with the situation? Not like react videos maybe (hope not), but with content that shows their interaction. Of course one could fake themselves like Milli Vanilli. But honest interaction.
E.g. with chess, isn't it like this: The game has been solved from a human vs computer perspective for a long time, yet at some point the content grew very fast when people had time to engage in all kinds of chess matters. So in the future, the product you pay may be not a composition but you composing, playing etc. I guess this does not contradict what enroe said about certain things per se becoming obsolete, but then who cares. If all you had to do is put an honest label on it or show yourself in some way or another. I don't get where the eccentric behaviour would come from.
And if the AI sound quality will be there at some point in the future, isn't it the artist that performs the good selection? Then how would anyone feel obsolete. Effort pays - I don't care much about what buttons exactly I need to press or what LLM knowledge I needed to learn. Sometimes discussions (not this one) about AI seem loaded with fear of loss to me. Then it reads like a discussion about who will get problems and who will not. That's not what I like about forecasting the future. I want to see it only positive, that's my honest take about this.

[enroe]

and, enroe to tell me the reason I don't find that argument supported is because I didn't read what you wrote is obnoxious. Tthe support for an argument being your argument is called begging the question.

If you think "classical music" or jazz will equally meet their doom, all this shows me is you have no idea what that music is or does. AI has no ideas. It goes by prevalence in what it's told to focus on and gives us a fake of it. Music from genuine ideas isn't subject to that.

The argument is essentially people that consume pop music amount to a group mind and lemmings will follow each other over the cliff. This happened with pop a long damned time ago. Copy after copy after copy, quelle surprise, finally the properly homogenous and bland rules OK.

Mmm... we've been through this before: Just because I outline a
development that you don't like, you can't simply negate that
development (i.e., you can, but...).

How about considering whether this development could be true?
I can only give a brief overview. The two relevant points are:

1. AI is not simply a large computer with a lot of database power.

2. AI development will continue rapidly. It will dominate and disruptively
transform the commercial creative sectors.

-------------------------------------------------------------------------------------------------

(1) We've already covered this. This isn't a hypothesis, it's a fact:

... is just a program running
... a parrot ... is not actually using language or speaking to you.

The AI introduces a completely new quality, namely "neural networks."
These have existed since the 1970s, but now CPU power and learning
capabilities are sufficient to apply "neural networks." And no calculations
are actually performed; millions of neurons are simply interconnected.
So neural networks function very similarly to the neurons in the human
brain.

For note-taking: AI can truly forge new connections, "new ways of thinking,"
just like the human brain. That is precisely the new quality that AI is now
bringing to the forefront.

-------------------------------------------------------------------------------------------------

(2) is a hypothesis, but a very, very plausible hypothesis:

Anything in the creative field that can be done cheaper and faster with
AI WILL be done with AI.

For music composition – in all genres, including rock, pop, ambient,
electronic, jazz, and classical – this means that companies releasing
new music will flood the market with AI-generated new compositions.
And they will thereby push all traditional human compositions to the
margins.

This applies to compositions!

So, the smoke-filled jazz club with old-school brass bands and
intellectual listeners will continue to exist. The classical concerts
performed by a state-funded orchestra will also continue. These
special events, of course, won't be rationalized by AI anytime soon.

But everything that is commercial, where costs and the resulting effects
are taken into account, will be AI-dominated. This applies to all new
compositions, publishers, etc.

Surveys already show that for the vast majority of listeners, it's important
that a song sounds "good". The fact that this song was created by AI is
of little interest to most.

User "Chagzuki" has raised an important point here: Due to the
dominance of AI in the commercial mass market – so goes his
hypothesis – music will become a kind of background noise for all
kinds of internet and entertainment content. I truly believe this is
very likely.

[Grizzellda]

i foresee a reprisal in local live music, small bands in small rooms...
people will still want to make music, even if only those people, ever go to other musicians gigs, it's still a niche market, ripe for exploiting!

I agree, how can AI replace that? No it just will not.

By the way vurt, I just recently noticed that you have over 110 thousand posts on KVR.

Have you ever considered...well I dunno, maybe considered...

Easing up a bit?

[enroe]

... That's not what I like about forecasting the future.
I want to see it only positive, that's my honest take about this.

Yes, I would also like a positive, optimistic outlook for the future.
But... mmh... on the other hand, this risks turning a blind eye,
failing to recognize real dangers, and thus also missing opportunities?

Perhaps some of the unpleasant consequences could still be prevented.
But if we close our eyes, if we ignore the negative effects, then we
have no chance at all. Right?

[Grizzellda]

... That's not what I like about forecasting the future.
I want to see it only positive, that's my honest take about this.

Yes, I would also like a positive, optimistic outlook for the future.
But... mmh... on the other hand, this risks turning a blind eye,
failing to recognize real dangers, and thus also missing opportunities?

Perhaps some of the unpleasant consequences could still be prevented.
But if we close our eyes, if we ignore the negative effects, then we
have no chance at all. Right?

The sense I get, is well...we have to just wade into the future, perhaps alot like wading into a waist high swimming pool! I have done that before!

Of course, the water gets deeper, and "can you swim"? becomes the question.