Jazz is like a safe haven

[dharma2020]

With all of the fake AI generated music (and video) appearing everywhere I'm really happy to just listen to old records made by humans.
Or just sit down and practice my instrument.
That's just something that generative AI will never do for me.

Anyone else feel the same? Curious about your opinions on this topic.

This text was generated with ChatGPT, ... JK

[mr. beaumont]

AI "art" is sad.

But if you look at how dumb we've become as a species, it's not surprising.

[omphalopsychos]

AI music to me isn’t that different than some manually written/performed commercial music. It serves the same purpose (low cost, high profit, low artistic depth). The only parameter changing is the cost is even lower. But music has been a B2C (business to consumer) model for decades. I don’t see a fundamental rupture in commercial music.

Jazz is nice.

[AllanAllen]

You're just running open arms toward dystopia aren't you? AI is the end man, we are losing whatever crumbs of factual information we had left, it's over. Anything and everything that the powers that be don't like will be deemed AI and we won't be able to check it.

Spotting AI images just got harder with Nano Banana Pro #AILiteracy - YouTube

But hey, at least the stocks go up...

[omphalopsychos]

Bro your worldview is sourced from tiktoks and you think AI is the problem?

[Doug B]

AI music to me isn’t that different than some manually written/performed commercial music. It serves the same purpose (low cost, high profit, low artistic depth). The only parameter changing is the cost is even lower. But music has been a B2C (business to consumer) model for decades. I don’t see a fundamental rupture in commercial music.

Jazz is nice.

So you're saying that Katy Perry is an AI Bot? The nerve! lol!

[DawgBone]

You're just running open arms toward dystopia aren't you? AI is the end man, we are losing whatever crumbs of factual information we had left, it's over. Anything and everything that the powers that be don't like will be deemed AI and we won't be able to check it.

Spotting AI images just got harder with Nano Banana Pro #AILiteracy - YouTube

But hey, at least the stocks go up...

It's all the same to me....it's the people who control the information. Whether it's the encyclopedia brittanica or google AI, the "facts" are determined by social engineers.

[Strat-itis]

You're just running open arms toward dystopia aren't you? AI is the end man, we are losing whatever crumbs of factual information we had left, it's over. Anything and everything that the powers that be don't like will be deemed AI and we won't be able to check it.

Spotting AI images just got harder with Nano Banana Pro #AILiteracy - YouTube

But hey, at least the stocks go up...

I don't see AI causing a crash worse than what we already went through with broadband and porn saturation in the 00s causing the collapse of public heterosexuality in the 10s. But it's possible I suppose.

We're in the future, the best we can do is adapt positively.

[AllanAllen]

It’s not about the art. It’s training on art, but the end use will be for propaganda. Socially, things are only going to get worse.

The point of the clip was AI is getting really good at fake photographs.

An unrestricted industry with this much potential power should be concerning to anyone.

[Strat-itis]

I actually agree that as technology advances that art gets more hollow because of how the music industry functions as a B2C like omphalo said. Shallower and cheaper art is produced to have a wider appeal and it controls the market. Rather than in the 20th century when music and movements grew organically and then were selected.

Tech also doesn't force people to be shallow with their use of it. But human nature when given convenience is to take the lazier option. So tech does have a tendency to kind of devolve people into shallower and stupider modes of living. While the minority will use it to advance themselves.

[omphalopsychos]

It’s not about the art. It’s training on art, but the end use will be for propaganda.

These models aren't being trained on art. "Art" makes up a tiny subset of that they are trained on. The reality is that 99% of human-produced content (i.e. the training data) is already commodity, not art. You're not protecting Wagner, Van Gogh, Bird, you're protecting [insert mega pop star here] and buzzfeeders/bloggers/influencers/content producers. This stuff is already commercial propaganda.

People saying don't introduce AI into art are the same people storyboarding, designing, filming, producing KIA commercials.

Art's still here. AI "art" isn't art. It's AI-generated "kitsch-as-commodity" and it sucks, but human-generated kitsch-as-commodity has already been a thing for a century.

[frabarmus]

Al has its positive applications, however the arts are not one of them. Al lacks five of the major building blocks of art that constitutes much of human existence; i.e. the five senses. No computer program can be written for the five senses. Not having the ability to hear, see, smell, touch, and taste is a major handicap to understanding the human experience. Without the basic understanding of humanity’s condition, how can art be made?

Here's AI's reply:

You’re raising a serious and long-standing philosophical objection, and it’s a strong one. At its core, your argument isn’t really about technology—it’s about what art is and where meaning comes from.
A few key points help clarify the debate.

---

1. You’re right about the senses—and embodiment matters

Human art is deeply rooted in embodied experience. Artists don’t just process information; they:

• feel pain and pleasure
• experience memory through smell and sound
• associate touch with comfort or danger
• live inside time, aging, loss, and anticipation

These experiences shape intuition, symbolism, and emotional weight. An AI does not experience a scream, hunger, love, or grief. It does not inhabit a body that can be harmed or cherished. In that sense, AI lacks what philosophers call phenomenal consciousness—the “what it feels like” aspect of existence.
So if art is defined as an expression of lived experience, then your conclusion follows: AI cannot truly create art in the same way humans do.

---

2. But art has never required the artist to directly experience everything

Here’s where the counterargument begins.
Human artists often create powerful work about things they have not personally experienced:

• war
• historical trauma
• other cultures
• even emotions they imagine rather than feel

They do this by:

• observing others
• absorbing stories, symbols, and traditions
• working within shared cultural languages

AI operates in a similar—but more extreme—way. It does not sense the world directly, but it is trained on human traces of sensory experience: language, images, music, and patterns shaped by millions of embodied people.
That doesn’t give AI understanding—but it gives it access.

---

3. The key distinction: generation vs understanding

This may be the most important point.
AI can:

• generate forms that resemble art
• recombine styles, motifs, and structures
• produce outputs that humans may find moving

But AI does not:

• intend meaning
• suffer for expression
• seek truth, beauty, or catharsis

In other words, AI can produce artifacts, but it does not participate in the human condition.
So when people say “AI makes art,” they are often confusing:

• the object (the image, poem, song)
  with
• the act (expression arising from a conscious, embodied being)

---

4. A useful comparison: tools vs authors

Throughout history, new tools have been accused of undermining art:

• photography vs painting
• synthesizers vs acoustic music
• digital art vs traditional media

What survived was not the tool, but human intention.
AI fits more cleanly as:

• a sophisticated instrument
• a mirror reflecting collective human culture
• a collaborator only insofar as a human frames and interprets the output

The meaning still comes from the human who chooses, contextualizes, or responds.

---

5. So can AI make art?

It depends on the definition:

• If art requires lived experience, intention, and consciousness ? no.
• If art is defined as objects that evoke meaning in humans ? AI can contribute, but only indirectly.
• If art is about human existence ? AI can imitate its language, but not inhabit it.

Your argument stands strongest when art is understood not as output, but as existential expression.

---

Conclusion

AI’s lack of senses is not just a technical limitation—it is an ontological one. It can simulate the surface of art, but not its source. What AI produces may be art-like, useful, or even emotionally effective, but the humanity in it does not originate from the machine. It is borrowed, filtered, and recombined from human lives.
So the deeper question may not be “Can AI make art?”
but rather: “Are we mistaking resemblance for essence?”
And that’s a question only humans can ask—because only humans are living inside the condition art tries to express.

[omphalopsychos]

I think we are safe.

[frabarmus]

I think we are safe.

Ouch!! That "greatest photograph ever" is so so so ..."chocolate boxy"!!

[AllanAllen]

These models aren't being trained on art. "Art" makes up a tiny subset of that they are trained on. The reality is that 99% of human-produced content (i.e. the training data) is already commodity, not art. You're not protecting Wagner, Van Gogh, Bird, you're protecting [insert mega pop star here] and buzzfeeders/bloggers/influencers/content producers. This stuff is already commercial propaganda.

People saying don't introduce AI into art are the same people storyboarding, designing, filming, producing KIA commercials.

Art's still here. AI "art" isn't art. It's AI-generated "kitsch-as-commodity" and it sucks, but human-generated kitsch-as-commodity has already been a thing for a century.

You aren’t understating my point.

It’s not about the art, the end goal isn’t art.

Generated photorealistic images will shatter our confidence in verifiable information and will be powerful tools for propaganda and the existence of these said images will make actual evidence harder to verify and easier to deny.

[omphalopsychos]

You aren’t understating my point.

It’s not about the art, the end goal isn’t art.

Generated photorealistic images will shatter our confidence in verifiable information and will be powerful tools for propaganda and the existence of these said images will make actual evidence harder to verify and easier to deny.

I understand your point. But real images and mass media have been doing exactly this for over a century. The problem you’re describing is not new, only the technical means are.

This is precisely what thinkers like Walter Benjamin, Walter Lippmann, Adorno, Debord, Chomsky, and Marcuse were already worried about. Benjamin argued that mechanical reproduction breaks the authority of the original and changes how truth is experienced. Lippmann showed that public opinions don’t relate to reality directly at all, but through mediated “pictures in our heads.” Adorno and Horkheimer saw mass media as producing conformity and weakening critical judgment. Debord argued that representation replaces lived reality entirely. Chomsky documented how media systems manufacture consent regardless of whether the underlying images are “real”.

None of these critiques depended on fake images. They assumed photographs, film, and broadcast media were real, and still concluded that truth would erode, propaganda would scale, and denial would become easier. AI imagery accelerates this, but it doesn’t fundamentally change the structure of the problem.

So the issue isn’t that generated images suddenly make verification impossible. It’s that modern societies already rely on trust in institutions, chains of custody, and interpretive frameworks rather than raw visual evidence. That was true long before AI, and it will remain true after it. The technology doesn’t create a new epistemological crisis from nothing.

[Strat-itis]

This AI ruining art thing is still based on the idea that you can't enjoy art if the art in the mainstream sucks. Ok? You have to be a follower to enjoy yourself in life? I didn't know that was doctrine. Why does the music you like have to be validated by the mainstream for you to enjoy yourself? The mainstream peasant folk songs probably sucked bad too..

[AllanAllen]

I understand your point. But real images and mass media have been doing exactly this for over a century. The problem you’re describing is not new, only the technical means are.

This is precisely what thinkers like Walter Benjamin, Walter Lippmann, Adorno, Debord, Chomsky, and Marcuse were already worried about. Benjamin argued that mechanical reproduction breaks the authority of the original and changes how truth is experienced. Lippmann showed that public opinions don’t relate to reality directly at all, but through mediated “pictures in our heads.” Adorno and Horkheimer saw mass media as producing conformity and weakening critical judgment. Debord argued that representation replaces lived reality entirely. Chomsky documented how media systems manufacture consent regardless of whether the underlying images are “real”.

None of these critiques depended on fake images. They assumed photographs, film, and broadcast media were real, and still concluded that truth would erode, propaganda would scale, and denial would become easier. AI imagery accelerates this, but it doesn’t fundamentally change the structure of the problem.

So the issue isn’t that generated images suddenly make verification impossible. It’s that modern societies already rely on trust in institutions, chains of custody, and interpretive frameworks rather than raw visual evidence. That was true long before AI, and it will remain true after it. The technology doesn’t create a new epistemological crisis from nothing.

Yes, exactly, the images(really, the AI system as a whole) are a tool to destroy tho institutions, chains of custody, and interpretive frameworks we rely on.

It’s a very powerful tool at that. And it’s puzzling why you’d be so welcoming to it when you understand the negative side.

Any productivity it creates won’t trickle down to us, it’ll be less employees with more workload and a tool that hallucinates results. The benefits of productivity increasing has historically never been freely given to employees. And with all the politics going into keeping AI unregulated I don’t see that changing here.

[omphalopsychos]

Yes, exactly, the images(really, the AI system as a whole) are a tool to destroy tho institutions, chains of custody, and interpretive frameworks we rely on.

It’s a very powerful tool at that. And it’s puzzling why you’d be so welcoming to it when you understand the negative side.

Any productivity it creates won’t trickle down to us, it’ll be less employees with more workload and a tool that hallucinates results. The benefits of productivity increasing has historically never been freely given to employees. And with all the politics going into keeping AI unregulated I don’t see that changing here.

Now it’s you who are misunderstanding my point.

You’re describing AI as a tool that destroys institutions, chains of custody, and interpretive frameworks. My point is that those institutions are not neutral safeguards of truth that AI suddenly corrupts. They are the systems that have always selected, framed, certified, and enforced what counts as legitimate evidence and belief. We’ve seen this repeatedly without AI. Think of the U.S. media consensus around weapons of mass destruction in Iraq, Israeli media coverage shaped by state security frameworks, or Russian and Western media each presenting radically different narratives of the Ukraine war. In none of these cases did institutions need fabricated images to control interpretation. They controlled it through framing, omission, repetition, and authority.

There is an irony here. You are worried about AI undermining institutions, but you are also treating those same institutions as if they were reliable anchors of truth before AI arrived. Much of the critique from 20th and 21st century thinkers is precisely that these institutions do not merely transmit facts, but actively control which facts are visible, credible, and actionable in the first place.

Your level of trust in institutions is almost quaint. You assume the state, the media, and official authorities naturally acted as benevolent guardians of truth prior to AI.

[omphalopsychos]

I think my point can be summed up very simply.


In art, AI produces commodities inside systems that already favored commodities. That photograph looks the way it does because platforms like Instagram already codified narrow standards of “good” photography.

In politics, AI produces propaganda inside systems that already favored propaganda. Before and without AI, states and media shaped belief through long term control of narratives and information flows. Synthetic pictures/videos are crude by comparison.


(And jazz is a safe haven, let's appreciate it.)

[DawgBone]

I think my point can be summed up very simply.


In art, AI produces commodities inside systems that already favored commodities. That photograph looks the way it does because platforms like Instagram already codified narrow standards of “good” photography.

In politics, AI produces propaganda inside systems that already favored propaganda. Before and without AI, states and media shaped belief through long term control of narratives and information flows. Synthetic pictures/videos are crude by comparison.


(And jazz is a safe haven, let's appreciate it.)

I watched a slew of angry hip hop/rap people ranting on X about AI generated beats being crap and ruining "real" artists a couple weeks ago but I couldn't help but recall how readily they embraced drum machines and samples and bashed anyone who got in their way about it as it rose to prominence by the mid 90's yet the reality was that was the death knell for a lot of session musician work. Now they get a taste! A truly perceptive person would've foreseen all of it with the advent of the synthesizer, or perhaps electric instruments in general!!!

[AllanAllen]

In politics, AI produces propaganda inside systems that already favored propaganda. Before and without AI, states and media shaped belief through long term control of narratives and information flows. Synthetic pictures/videos are crude by comparison.


(And jazz is a safe haven, let's appreciate it.)

I hope you’re right, but I don’t think you will be.

[omphalopsychos]

I hope you’re right, but I don’t think you will be.

I know. Could you imagine if our politicians STARTED LYING TO US? THAT WOULD BE A DISASTER.

[omphalopsychos]

I watched a slew of angry hip hop/rap people ranting on X about AI generated beats being crap and ruining "real" artists a couple weeks ago but I couldn't help but recall how readily they embraced drum machines and samples and bashed anyone who got in their way about it as it rose to prominence by the mid 90's yet the reality was that was the death knell for a lot of session musician work. Now they get a taste! A truly perceptive person would've foreseen all of it with the advent of the synthesizer, or perhaps electric instruments in general!!!

A P Carter was already doing something similar in the 1920s. He traveled through Appalachia and heard songs sung by ordinary people. These were shared folk songs with long oral histories. He memorized or wrote them down then brought them back arranged them and recorded them. He copyrighted them and made money from them. I'm saying this as someone who enjoys both country music and hip hop.

[djg]

You’re describing AI as a tool that destroys institutions, chains of custody, and interpretive frameworks. My point is that those institutions are not neutral safeguards of truth that AI suddenly corrupts. They are the systems that have always selected, framed, certified, and enforced what counts as legitimate evidence and belief. We’ve seen this repeatedly without AI. Think of the U.S. media consensus around weapons of mass destruction in Iraq, Israeli media coverage shaped by state security frameworks, or Russian and Western media each presenting radically different narratives of the Ukraine war. In none of these cases did institutions need fabricated images to control interpretation. They controlled it through framing, omission, repetition, and authority.

the difference is that AI tools like veo or nanobanana have "democratized" fake news. they are also prime stalking tools that will change our societies for the worse. and the actors described above now have tools at their disposal that are the wet dream of any dictator.