Josh has a few thoughts about AI…

[DaBigBadaBoom]

 

[DaBigBadaBoom]

There's a theory I read - kinda like the dead internet theory - which stated that, because AI/LLMs are trained in humanity, it will inherit our worst traits.

And that's why so many people - myself included - despise it. It's like the worst people I know: dishonest, clueless, egomaniacal... Dunning-Kruger writ large.

 

[CamJazzBass]

There's a theory I read - kinda like the dead internet theory - which stated that, because AI/LLMs are trained in humanity, it will inherit our worst traits.

And that's why so many people - myself included - despise it. It's like the worst people I know: dishonest, clueless, egomaniacal... Dunning-Kruger writ large.

I remember reading about AI and someone asked, how will we keep it from killing us all, just going nuts. The smart scientists said, well, we will for sure code it with our morals, kind of like an Asimov thing. I read that and thought, yep, we'll be dead for sure if they do that.

 

[CamJazzBass]

All the people that work in tech, I can't believe they are not going rogue. I mean, they are basically training their replacement. I would think they would be smart enough to figure that out and low key be sabotaging the AI from the inside.

Dadgummit man, I just can't seem to ever get this AI to work right. I don't think it will ever be any good (wink, wink)

 

[CamJazzBass]

They did. And thought it was a great idea.
There's even an Australian (I think) company developing AI driven armed robots and drones named Skyborne (I guess Skynet was already trademarked).

Sci fi authors have been forward thinking and predicting the worst things from AI for decades. Yet, we have scientists who just can't wait to do it. I don't get it. They are even trying to make the robot dogs like the ones in Black Mirror. That is nightmare fuel. What good could possibly come from such an invention?

Black Mirror:

Real life:

 

[DaBigBadaBoom]

All the people that work in tech, I can't believe they are not going rogue. I mean, they are basically training their replacement. I would think they would be smart enough to figure that out and low key be sabotaging the AI from the inside.

Dadgummit man, I just can't seem to ever get this AI to work right. I don't think it will ever be any good (wink, wink)

Hyper-specialisation in education has created all these groups which can't talk to or understand each other. That creates a notion of intelligence which is illusory.

Most tech workers are middle and upper middle class people whose privilege leads them to believe their crap behaviour won't effect them.

No one is gonna rescue us from the inside. We have to rescue ourselves.

But we won't.

 

[BioDriver]

I have been extremely vocal about my skepticism and distrust of AI, both in personal and corporate settings. My company's solution was to promote me to a strategy position to act as a devils advocate and counter to the hype.

These people are shoving it down or throats whether we want it or not. We can't close this can of worms, but they don't seem to care about containing it or being ethical about its use

 

[CamJazzBass]

I have been extremely vocal about my skepticism and distrust of AI, both in personal and corporate settings. My company's solution was to promote me to a strategy position to act as a devils advocate and counter to the hype.

These people are shoving it down or throats whether we want it or not. We can't close this can of worms, but they don't seem to care about containing it or being ethical about its use

They encourage us to try out the AI we have and use it in out work. I'm not training my replacement. It is coming whether I like it or not, but I'm not going to take a part in helping it along.

 

[DaBigBadaBoom]

I have been extremely vocal about my skepticism and distrust of AI, both in personal and corporate settings. My company's solution was to promote me to a strategy position to act as a devils advocate and counter to the hype.

These people are shoving it down or throats whether we want it or not. We can't close this can of worms, but they don't seem to care about containing it or being ethical about its use

Too big to fail is a belief not a reality... By which I mean, the world has invested 2.5 trillion in AI. And destroyed all notion of intellectual property.

Compared that to the subprime housing crisis, where the value of the property involved was 600 million.

If the markets all decide that AI is essentially the next metaverse, no amount of bailout will fix it. So global recession on a scale not seen since the great depression.

That's if it fails.

If it doesn't fail it's promise is to destroy all paid human labour.

During the great depression the unemployment rate was 25%.

Curently the TRU - which is the the True Rate of Unemployment ie "the percentage of the U.S. labor force that is functionally unemployed" is 23.6%.

If AI reduced employment by even 10% you're talking about a catastrophic contraction of employment.

So.

Damned if AI succeeds, damned if it fails.

And 10% is nothing compared to what they're promising, which is 90%+ reduction in employment.

All of which assumes we haven't spent 3 trillion to create a machine which wipes out all life on the planet.

 

[BioDriver]

Too big to fail is a belief not a reality... By which I mean, the world has invested 2.5 trillion in AI. And destroyed all notion of intellectual property.

Compared that to the subprime housing crisis, where the value of the property involved was 600 million.

If the markets all decide that AI is essentially the next metaverse, no amount of bailout will fix it. So global recession on a scale not seen since the great depression.

That's if it fails.

If it doesn't fail it's promise is to destroy all paid human labour.

During the great depression the unemployment rate was 25%.

Curently the TRU - which is the the True Rate of Unemployment ie "the percentage of the U.S. labor force that is functionally unemployed" is 23.6%.

If AI reduced employment by even 10% you're talking about a catastrophic contraction of employment.

So.

Damned if AI succeeds, damned if it fails.

And 10% is nothing compared to what they're promising, which is 90%+ reduction in employment.

All of which assumes we haven't spent 3 trillion to create a machine which wipes out all life on the planet.

We're essentially in stagflation and AI failing would push the US into a depression while triggering another global recession. And that's the best case scenario. The AI corps are only held up by speculative growth that is overshadowed by their insane debt

 

[-Asdfgh-]

Also, if you're asking it questions about music gear, it will mostly pull its data from forums and reddit.

If everyone here decided to post enough times that the Fender Jazz Bass was invented by Miles Davis, there's a good chance it would eventually claim that as fact.

Back in the first days of Wikipedia I constructed an elaborate but fake biography (photos, etc.) for a fictitious early soccer player called Neville Hill. Neville Hill is actually rail repair depot near Leeds. I wanted to see if it would become accepted. But I had a pique of conscience and deleted it.

 

[DaBigBadaBoom]

We're essentially in stagflation and AI failing would push the US into a depression while triggering another global recession. And that's the best case scenario. The AI corps are only held up by speculative growth that is overshadowed by their insane debt

But don't worry, our hypernormal world will find a rational solution to it all, which will benefit everyone.

 

[jbybj]

AI is pretty good and predicting results for such questions as "What bass frequencies should I bump/cut for the best IEM experience." Some of you do not need that kind of thing, others need a place to start.

I humbly, wholeheartedly, disagree with this. When it comes to tweaking gear, doing, flailing, and failing is infinitely superior to being told what to do. Not just AI, but also random internet humans being asked, "How should I set my big muff pi?"

 

[jbybj]

Fact: Auto mechanic Larry Graham invented the Jazz Bass.

 

[Enrique M]

I humbly, wholeheartedly, disagree with this. When it comes to tweaking gear, doing, flailing, and failing is infinitely superior to being told what to do. Not just AI, but also random internet humans being asked, "How should I set my big muff pi?"

Maybe you have never seen me flail? but I get it.

 

[5StringBlues]

Back to the oceans guys.