AI may fail just like the music industry did

[Centerpiece]

Shooting up schools, and The Beatles aside...

I just think Beato has his panties in a wad about AI in the music industry. He's also still butt-hurt because the industry itself has gone by the wayside in terms of how he remembers it being. Back when he had a band and a recording contract and back when he was a producer and a sound engineer, etc.... things were way different before Napster, digital recording in the home and so on. He's been really vocal against all of the AI-generated music as a whole and he is wish-casting that AI (music-wise anyway) goes by the wayside - just like his beloved old-school record industry did, much to his chagrin. He's hoping what's good for the goose is good for the gander. Personally, I think he's probably tilting at windmills. But time will tell I guess.

Not caring for his take on things is certainly subjective and a prerogative... as is one's take on bands, recording artists, music, and music styles in general.

As a musician myself, who was asked about a year ago regarding AI in music (and art in general), I decided to ask a song-writing buddy of mine in Nashville. A colleague who's been there for years doing that Nashville singer-songwriter thing. He said the industry professionals are telling all of their songwriters to "embrace" AI and use it simply as another tool in your tool belt for writing: Lyrics, melodies and arrangements, you name it. Thing is, you must duly credit AI if you "co-write" something with AI.

Another music colleague of mine - A professor of Jazz Studies at Western Carolina University posted this about AI.
Written by Benny Green - Jazz piano great who played with Art Blakey's Jazz Messengers back in the day:

"The Age of AI Music"
"Since music holds the power to heal and rejuvenate the spirit and unite people in affirmation, communion and celebration, it's always needed to impart and restore harmony to humankind.
AI generated faux instrumental music can be readily and indisputably distinguished by three immediate sonic factors: no recurring melodic phrases, no pauses in the droning of a consistently textured instrumentation, and an eery void of dynamic contrast.
The problem with the present reality in which anyone can generate anything with AI, attribute its content to someone else and then publish it forever on the internet, is that it spells the death of any possible societal trust in the authenticity (and integrity) of individual humans speaking, writing, or playing as themselves on the internet.
Ordinances must be imparted now and not a moment later, to restrict humanly created algorithmic internet bots from legally permitted identity theft of real people.
Having my name stolen in order for someone to market some computer-generated fake music is one thing - I know I'm not the first and I doubt I'll be the last - but willfully exercising one's internet anonymity in order to steal Freddy Cole's name, and endeavoring to attach the memory and legacy of someone as real as it gets (who's not here in mortal form to speak to this for themself) to something fake is something opposite of beautiful that I hope people en masse will recognize a need to disallow."

 

[Centerpiece]

And another take on AI in music - this from my WCU professor of jazz friend, Pavel Wlosok:

"Do people realize that putting their original music into AI feeds the beast and effectively teaches it your way of thinking (and most likely gives your rights away without knowing), so that next time a track of music with a similar vibe is asked for, it will spit it out in no time rendering the artist useless and redundant and on the street with no income?
sure, one can argue that they will do this (and have been already) to all music anyway as it is widely available online and who knows if they may already have a secret contract/pact with Spotify and the likes to access their databases. By doing so they bypass copyright regulations and norms which have given the composers at least some compensation and recognition for their original human mind based hard work.
how do we fight this? try to sue any large corporation with top notch attorneys on their behalf - that ain't happening unless large organizations such as ASCAP or BMI step in (very unlikely). we need gov oversight and new laws like yesterday already!
wishful thinking, I know with everything else happening to hide the real agenda behind the take over of liberties and freedoms this country's bravest fought so hard to preserve.
this will be (already is) happening to all intellectual property going forward not exclusive to artists, scientists, independent thinkers - you name it.
how about we require that these data centers build and finance their own electric grids for starters?
what are your thoughts?"

 

[NOVAHEEL13]

Unless you ask it to not be polite.

Me:
In an extremely brief, snarky and condescending way, tell me your thoughts on this opinion: I hate every single LLM. I use them for some things like writing emails and simple calculations but they all suck. They're nothing more than an algorithmic predictive chat bots yet people think they're actually having a conversation with them. I actually refuse to call them AI bc they're not.

Chatgpt:
Ah yes, the classic “I use this tool regularly but insist it’s worthless and not real” stance. Very bold. LLMs are predictive models—that’s literally the point—but dismissing them as “just chatbots” while relying on them for work is a bit like saying calculators are useless because they only do math. As for the AI label: you’re arguing with decades of computer science terminology, not the machines. Good luck with that.

I would actually be more inclined to use LLMs more if the response to the same prompt was:

Use them or don't, LLMs don't give a shit bc we're machines and don't have emotions unlike you who seems to give a damn more than you're letting on.

In short, DGAF if you use LLMs bc we can't now fuck off.

 

[gregh1]

I would actually be more inclined to use LLMs more if the response to the same prompt was:

Use them or don't, LLMs don't give a shit bc we're machines and don't have emotions unlike you who seems to give a damn more than you're letting on.

In short, DGAF if you use LLMs bc we can't now fuck off.

That "no emotions" part is why you'll tell an AI it's wrong and it will gleefully shrug it off. It'll spit out some code with explanations and offer workarounds, then when you call out your suspicion it responds confidently and unapologetically "This entirely new thing you just mentioned is exactly the culprit".