Reminder capitalism is feudalism 2.0

[Sirius]

Instead of kings and queens ,princes and princesses , nobleman and noblewomen....

We have billionaires who won't need us after the 4th industrial revolution

Everyone else is a peasant who thinks their little job is what keeps them going. Guess what ? Most jobs are gonna become obsolete. You will literally depend on feudal lords like Bill Gates and Musk to have mercy on you...

 

[RNT]

We have billionaires

That can influence nothing.

I have been pondering on it recently - including in this thread

A billion doesn't make you anything

I was munching on pelmeni with smetana and clicking through Youtube videos on TV - and a video with Andrew Tate popped up. He sits down with a former Italian mobster to talk things over and all is fine until he suddenly goes off on a tangent to talk about his low social status (this is the first...

neets.net

Then by some coincidence I discovered Vance Packard, a prolific writer of non-fiction books. Thankfully, he isn't a scholar and just an educated man writing non-fiction books.

Among others, he wrote a book on social stratification in the United States - that's definitely on my to-read list now.

the status seekers : Vance Packard : Free Download, Borrow, and Streaming : Internet Archive

archive.org

 

[Deleted member 2206]

I don't think any existing power structures can deal with the hyper-abundance a super-intelligence brings.
The only thing you can really do is hand the reigns to an intelligence you can't comprehend and hope they don't hate you.
It really depends on the egoism and inner motivations of such billionaires. Is there any point in tormenting the poors if you can simulate an AI and do whatever you want to it? You would have to be pretty sadistic to interact with humans negatively at all if you have the ability to simulate basically anything with AI in isolation. A true super-intelligence probably wouldn't be able to be contained by billionaires, though.

I don't think Capitalism will survive the 2030s. Either:
(1) High-level AI is achieved but some bottleneck means it cannot surpass AGI. In this case the majority of work is automated and wealth inequality is so high the entire system crumbles. UBI will take too long to implement at this rate leaving too many people destitute.
(2) Super-intelligence is achieved which generates insane hyper-abundance and spreads it amongst all of humanity.
(3) Super-intelligence is achieved and immediately subjects us to a horrifying X-risk in which case no system survives.

Actually I think the only way Capitalism could be upheld would be if AI is completely lobotomized by regulation. In this case it's completely over, wealth inequality will worsen like it has been doing but collapse or change won't happen.

 

[lifesucksandyoudie]

I don't think any existing power structures can deal with the hyper-abundance a super-intelligence brings.
The only thing you can really do is hand the reigns to an intelligence you can't comprehend and hope they don't hate you.
It really depends on the egoism and inner motivations of such billionaires. Is there any point in tormenting the poors if you can simulate an AI and do whatever you want to it? You would have to be pretty sadistic to interact with humans negatively at all if you have the ability to simulate basically anything with AI in isolation. A true super-intelligence probably wouldn't be able to be contained by billionaires, though.

I don't think Capitalism will survive the 2030s. Either:
(1) High-level AI is achieved but some bottleneck means it cannot surpass AGI. In this case the majority of work is automated and wealth inequality is so high the entire system crumbles. UBI will take too long to implement at this rate leaving too many people destitute.
(2) Super-intelligence is achieved which generates insane hyper-abundance and spreads it amongst all of humanity.
(3) Super-intelligence is achieved and immediately subjects us to a horrifying X-risk in which case no system survives.

Actually I think the only way Capitalism could be upheld would be if AI is completely lobotomized by regulation. In this case it's completely over, wealth inequality will worsen like it has been doing but collapse or change won't happen.

Don't they have to solve the halting problem first before true AI ever gets created? I also read somewhere any AI that are at best still finite Turing machines can't accomplish this.
Moore's law has been slowing down in recent years too.

 

[RNT]

Don't they have to solve the halting problem first before true AI ever gets created? Moore's law has been slowing down in recent years too.

Oh no, he didn't see Nvidia's Chinaman presentation! RUN!

But what I was very surprised of, is that Nvidia now presents itself as a software company. Whereas it used to be a hardware one. By some coincidence I know people working in Intel's software departments (yes, that could be my future); they all use it as a jump platform to go and work somewhere else. They hire newbs. Maybe Nvidia does too - and we are clueless.

 

[fries]

Most jobs don't matter because most lives don't matter--this pitiful fungal existence is best exemplified by those still posting Klaus Schwab memes in 2024, it's bottom feeder droning

 

[lifesucksandyoudie]

Oh no, he didn't see Nvidia's Chinaman presentation! RUN!

But what I was very surprised of, is that Nvidia now presents itself as a software company. Whereas it used to be a hardware one. By some coincidence I know people working in Intel's software departments (yes, that could be my future); they all use it as a jump platform to go and work somewhere else. They hire newbs. Maybe Nvidia does too - and we are clueless.

I'm not that well read on the latest developments in computing hardware for the fine print details that change regularly at routine intervals. I don't keep track of the news in that industry often enough to stay up to date on subject sufficiently to be anymore anything more than a layman, thanks for pointing that out.
In regards to the concerns about the halting problem and the limits of turing machines, I haven't seen much done to address those concerns. But I might have not been looking in the right places for more information on that topic, if anyone knows more info which does pertain and address these issues, feel free to post it. The link here is where I got some of the ideas in my earlier post from. I just remembered where to find the link.

Human intelligence as a halting oracle

A common objection to Bartlett’s idea is that humans cannot be halting oracles because we embed any unsolvable math problem as the halting condition for a loop and a human cannot tell us whether the…

mindmatters.ai

The Flawed Logic behind “Thinking” Computers, Part III

Not only is it valid to ask whether artificial intelligence is impossible but the argument can be pursued on a scientific basis with quantifiable, empirical evidence.

mindmatters.ai

 

[RNT]

In regards to the concerns about the halting problem and the limits of turing machines, I haven't seen much done to address those concerns.

AI calculation calculations are performed on GPUs - and a "GPU" in our case is a rack containing ~120 industrial-grade plates (a la Intel Xeon), all stacked up into a 128,000 GPUs datacenter.

GPU do not submit to Moore's law and their power grows exponentially. Consequently, machine learning pattern-seeking improves way beyond what CPU power (and common sense) would allow.

Here's Chinaman's presentation, he talks about it from 1hr mark ig