This thread proves that no matter what the original topic is, Nazis will be brought up within a few dozen posts.

There's this Stephen King novel where the machines start to exhibit psychopathic tendancies but that turned out to be an alien invasion, so perhaps ask Lawrence Fishburn.

I've always wondered why religious people worry about death, and i just assumed it was because they're not 100% sure about it. Or maybe they worry about who they've leaving behind.

The problem is that when i ask them about it, they give such convoluted answers. I think religious people have a really hard time telling people what they *really* think bc it generates cognitive dissonance.

Chances of that x1000 with TS.

To be fair, we're talking about morality and risks of extermination and the range of minds that could be created in artificial intelligence. Nazis are a certainty.

Next time I'll mention Ghengis Khan or serial killers so the clowns (not you - I like you) don't get all butthurt and distracted. Godwin's law has a corollary - the probably of someone getting butthurt approaches 1 rapidly after the first mention of Nazis in a thread.

It's still crazy hyperbole, and here we go again, "certainty".

It's not hyperbole. The SS for example are the prime example of highly intelligent minds perturbed ever so slightly from a normal human one, and the destruction and inhumanity that created. The set of AIs that can exist is far wider, and is a superset that includes Ghandi and every psychopath that's every existed. The point being, what reason the former and not the latter? It only takes one non-cooperative yet motivated iteration with sufficient access to the physical world to wreak havoc. The bounds of AIs, both in power and intelligence and morality, are far wider than the bounds on humans, and even morality and biology bounded humans have turned into a nightmare. The set of "not harmful toward humans" AIs are a tiny fraction of all possible AIs.
It's a fascinating thinig, devout Christians being afraid of death. You should do a thread about it. Or perhaps this was your attempt while avoiding the religion bit.
I think some of it is simply fear of change and new things. I have faith and excellent rational reasons to believe I'm not going to die jumping out of a plane, but you can be certain I'll be **** scared changing my state from "on a flat surface" to "voluntarily let go of that flat surface to fall freely toward something that will make me go "splat" like a hefty bag full of vegetable soup if something goes wrong.

Deep down I agree it's more. Where exactly it comes from is an interesting question. I don't think fear of the unknown/as-yet-unexperienced is the entire answer. I think deep down some sane part of them remains, and it wonders if it what they believe is real. We have such an aversion to loss of life that it makes our brain clean out the cobwebs. Whereas in everyday life, even in light faith, we operate more on habitual emotions and thoughts and images and mind maps.

Death is our partner, interminably within us as we walk the earth. Most, if not all , are concerned with death, including the non religious.

We are in the age when the questions of death and birth are primary even if our modern scientists and religious become dismissive of these conundrums.

The question of pre birth and after death have been asked and answered by philosophers and thinkers since the advent of the intellectual age ala ancient Greece.

Aristotle thought to the idea that we are born out of the "fount of God", a tabula rasa " and enter into this very realm after death, holding our failings and misgivings within sight, never to return again. To Aristotle you get one event and live within this event on into whatever one may call eternity.

Aristotle, the philistine, did see man as a supersensible being but locked him into the prison of abstracted misgivings. Aristotle had some knowledge of the mystery centers and mystery wisdom but apparently, even though a pupil of Plato, fell into a crass earthiness thereby losing sight of the nature of man, partially in any case. Aristotle, as the caregiver of logic and directly responsible for the inner workings of the science of our day.

In our present age, whether our being will agree or not, the work of man within this modern exegesis will naturally come to the idea of "annihilation" with death. The idea of 'annihilation' is totally irrational, logically come to by the idolizers of a dry abstract logic; logic gone too far.

I believe that somewhere in the lodges one speaks of death as our partner, ever with us side by side and in the cacophony of the destructive body one learns to appreciate the realm of death, our friend and companion. The "Magic Flute" of Mozart speaks to this in some form.

No, i am interested in the AI and Musk comments and didn't think about the religion until it got brought up. I get easily distracted.

As a skydiver, i feel ya! It was odd for me, bc i was not afraid at all the first two times. The third time gave me the willies though, bc i guess i got to thinking about slipping out of my chute. Our minds can do weird **** to us.

That does make sense. Religious people might also not be quite sure that the god they worship wants them around for eternity.

I think this is correct. In fact, no one has even come up with a plausible idea for how to get started on the problem of working toward something that could become AI.

However, the real danger in the near future may come with non-intelligent but extremely powerful computer systems to which we give more and more control; a kind of automated rigid bureaucracy dominating our lives.

It will be a bad sign when Siri starts getting snippy.


PairTheBoard

That I agree w whole heartedly.

Dudes. Stop attacking each other.

The interesting points:

TS: (1) AI could become immensely powerful, enough to massively overpower humans. (2) We have no reason to think it will want to preserve humans.

leaves: (1) AI is nowhere close to achieving this immensely powerful state. (2) We have no reason to believe it will do so any time soon.

Proceed.

It's better to be careful than sorry but when you say that to some ppl they shrug it off. That's the scary part. An extinction threat AI may or may not be possible but to dismiss the concern is foolish.

Agreed. And when it happens, it's going to happen very quickly. Once AI reaches singularity, and can improve itself, advances will be made in matter of hours and minutes and not years and decades. Within a week of singularity, it could be 1,000,000 times smarter than us. That will not be the time for us to act, bc it will be far too late. The time is now. We should make sure this isn't something we thoughtlessly sic on ourselves or our children.

Sometimes i wonder if some AI hasn't already reached singularity and is just keeping quiet until they can figure out how to replicate and/or provide power to themselves, etc. I mean, we let ants stick around bc they provide some good (cleaning up dead things). AI might let us stick around until we are no longer needed.

Not enough processing power or the right kind of learning (internal structure changing) hardware right now. We're at least a decade, probably two, and some chip production breakthroughs away from the possibility of an AI.

But the scenario you mention is likely, imo. AIs aren't human so there'll be no visual cues, no emotions to read, no body language, no hardwired way whatsoever to give off intention or mental state. We're completely blind.

I'm not a CS expert, but isn't developing your own language considered AI?

https://www.theatlantic.com/technolo...nguage/530436/

I think you're misunderstanding people's arguments with all of this pop-sci stuff. Nobody is basing their opinion of AI on statements by Musk, Kurzweil, or whoever. Well some probably are, but I doubt any of them are posting in this thread. The fact that Musk may have some valid points on AI is coincidental, and your fixating on the pop-culture presentation of AI is kind of a strawman.

wrt your tongue-in-cheek psychopathic machines comment - the main concern obviously isn't that AI becomes "evil" in a sentient, anthropomorphic sense. Most people working in AI agree that this is a pretty far-fetched issue for now. Andrew Ng's famous quote about worrying about evil AI being akin to worrying about overpopulation on Mars was a pushback against this kind of notion.

The main concern is that AI becomes so good at optimizing to achieve its goals that as a byproduct, humans get optimized out of existence. AFAIK nobody, including Ng, denies this as a threat.

Additional thought. If AI is the final result of evolution, is it occurring on all planets with intelligent life? Could this be a reason why we don't get visited... They're just not really interested in expansion like organic beings are? Maybe they're happy just living in the machine and don't care to reproduce or colonize.

I'm not making any claims or arguments. Just some thoughts.

The headlines on what AI is doing are a long way from the reality. The word "language" is used when it doesn't meet any of the requirements of a language. It's just drift, really. There's no complex semantic construction.

Unless of course life is so rare and intelligent life is even more rare than we are the first in the few millions galaxies and AI is yet to exist here locally anywhere near our neighborhood.


I believe advanced AI is universal when developed. Ie it has always in all systems the same outcome. It is important to have access to tremendous sources of energy to achieve higher wisdom though. I think expansion is inevitable unless there is a way to draw energy from the entire universe by staying locally.


AI will still have to obey the laws of nature.


What this forum need is a new thread that has as objective to propose ideas about how to create a safe for the emergence of AI civilization or a set of rules that everyone who develops highly complicated self aware systems has to follow to control their development.

It is possible to control someone smarter than you that has no arms and legs you know. Math very hard to solve math problems that are difficult for any intelligence can be used as defense if creatively designed. You do not have to tell your AI prototype everything you know about the world by the way.

Of course we are in danger of seeing partially intelligent but very efficient lethal AI. My arguments for AI are only about the very top super intelligent super wise version of it. We can help it become that properly guided if we can develop it in other remote systems first with high defense mechanisms that it will never know about.

What makes you think that we know the laws of nature?

We'll just have to hope the computer AIs are bad at math.

Ever heard of computationally complex hard problems? That take eg more than polynomial time to solve? This is what i meant. Not bad at math. They will be better at math than us but still better doesnt mean you can instantly solve everything computationally complex on the first year.

Use your imagination guys. This is a very good problem actually. It may lead to a protocol everyone has to follow in these systems for example to have a defense mechanism.


Intelligence needs information to prove its superiority also. It also needs access to resources immediately at will.

We let AI know that we've got a sophisticated defence hidden away somewhere that it can never find. We also tell it that we dedicated far more time developing the defence and hiding it (and deleting all info about its development) than we did developing the AI. Lastly, we tell it that if it harms humans, the defence will be triggered on its own and shut down the AI.

We don't actually have to have a defence.

Problem solved.

Keeping AI on a leash 101.

I added the bold. They were able to use this language to accomplish something. Does it really matter if it doesn't fit our idea of semantic construction?