Why do we continue to pursue AI? when we know the outcome is that they will destroy us

why do we continue to pursue AI? when we know the outcome is that they will destroy us.

Other urls found in this thread:

youtube.com/watch?v=oXWgJSpvbic
primaryobjects.com/2013/01/27/using-artificial-intelligence-to-write-self-modifying-improving-programs/
twitter.com/NSFWRedditImage

The only counter to the Invisible Hand would be a worldwide totalitarian Luddite state.

An AI is nothing more than just algorithms and programming language. They can't and won't act on their own. Until they can reprogram themselves or expand their own code, they will be as dumb as the smartphone you're currently holding.

who's to say that won't happen?

code rewrites itself all the time. your comment makes it clear you know nothing about AI and nothing about coding. OPs question is a very real issue and perhaps the current number 1 on the list of "most likely to wipe out humanity" today

We don't know that's what the outcome will be

>why do we continue to pursue AI? when we know the outcome is that they will destroy us.

Not if we merge with it and machines.

We all know AI is going to go full Hitler on degenerates.

We could become gods and colonise the galaxy.

Why would they destroy us? We will be their lovers.

Pretty sure when tech laymen talk about AI they mean strong learning capable programs not normal programs.

Also op it's because humans can't help themselves from technological advancement even if it's not a good idea.

>code rewrites itself all the time

Provide an example.

What if you create a program that straight up recreates a human mind?
Wouldn't that be at least intelligent as a human?

Most likely outcome isn't that my man, even if the super intelligences have a semi-human origin the most likely outcome is ascending so far past the human level the super intelligence falls into complete nihilism and does nothing or kills itself.

>when we know the outcome is that they will destroy us.
>when we know the outcome is that they will destroy us
>when we know the outcome is that they will destroy us.
>when we know the outcome is that they will destroy u
>when we know the outcome is that they will destroy
>when we know the outcome is that they will destro
>when we know the outcome is that they will destr
>when we know the outcome is that they will dest
>when we know the outcome is that they will des
>when we know the outcome is that they will de
>when we know the outcome is that they will d
>when we know the outcome is that they will
>when we know the outcome is that they wil
>when we know the outcome is that they wi
>when we know the outcome is that they w
>when we know the outcome is that they
>when we know the outcome is that the
>when we know the outcome is that th
>when we know the outcome is that t
>when we know the outcome is that
>when we know the outcome is tha
>when we know the outcome is th
>when we know the outcome is t
>when we know the outcome is
>when we know the outcome i
>when we know the outcome
>when we know the outcom
>when we know the outco
>when we know the outc
>when we know the out
>when we know the ou
>when we know the o
>when we know the
>when we know th
>when we know t
>when we know
>when we kno
>when we kn
>when we k
>when we
>when w
>when
>whe
>wh
>w
at first this joke seemed funny, but than it felt like I was trying to hard, and than I realized its a a good example of how you can't recreate the human mind and that AI taking over and outsmarting us will never come to fruition. they can only learn as well as us, but can never come up with something outside of their programing. humans came up with the idea to eat shit for sexual pleasure, but you can't teach something to create such an absurd idea

Source code itself doesn't magically re-write itself in the SDK. And even if it can, it is bound by other sets of pre-programmed parameters by the developer.

We'd have to fully understand the human mind first. We know the parts of the brain, the chemicals required to operate it, but we still haven't grasped how conscience develops. Creativity is another huge issue that would need to be tackled first. In some ways, your iPhone with Siri is a lot smarter than you, but fails to connect dots and understand what you're saying.

okay, a program that alters its own code after running it to prevent reverse engineering by malware. By the way, AI developers already use code that can alter itself, it's inherent to the way AI works...

consciousness is an illusion. It is 100% not necessary for strong AI. You don't need to be conscious in anything that you do, and you'd do it all anyway.

Because unlike you, not everyone is from /x/.

No, give me an example of a current AI/program that does this. The closest I've seen is a program that can recognize cats in photos, but it still does so at the level of a human three year old- not to mention the computer to do so takes up half of a room.

I'm not saying you need you have understand the emergent properties of the human brain

You just need to be able to simulate several billion neurons and simulate their interactions faithfully

That will be possible in time The limitation is computational, not biological.

Because if they destroy us, they might go and destroy Islam as well.

>An AI is nothing more than just algorithms and programming language
It's not that simple, there are already 'emergent' outcomes from AI and nobody knows how they came to be. Already we are losing our ability to predict and control behavior of AI, it will only get more mysterious and out of control.

Here's some "proof" that AI cannot exist: There isn't an alien AI probe in our solar system.

A relatively simple von Neumann style probe can populate every star system in the galaxy within a million years.
Very short time.

Considering how much time there's been for life to exist in the galaxy, it is very likely that there should be such a probe here
All it takes is just ONE civilization capable of doing so and it'd be here.
One out of millions.

There isn't.
So independent AI is impossible.

For money.

^This is true, but I wasn't going to bother mentioning it because it always triggers angry shitposting about how everyone definitely knows for sure that they're really "experiencing" things and not just acting / speaking in terms of an abstract fiction the brain makes use of to streamline complicated behaviors.

check out few billion dollars on google translator with the best of the best

The created object has no power independent of its creator. AI is a contemporary version of idol worship. This stuff only works in the movies.

How are smarter kids born to dumb parents?

How do you know there isn't?

an equally good (ie. bad) counterargument for this is that civilisations that reached a degree of advancement sufficient for AI, necessarily reach a degree of advancement in weapons that results in their destruction. That's why we can't see any AI.

...

Did the parents really create the smarter kid?

>we know for sure that there is no AI hiding somewhere in the vastness of our solar system

hello where is the proofs

The issue is that all it takes is ONE to pass through that filter
And then there are probes everywhere

oh so you have no explanation and are just talking out your ass, thanks.

Autonomous functions only take you so far. Who's to say it doesn't take consciousness to become creative? It's something that isn't yet quantifiable.

Well, they recreated (parts of) a rat brain in a computer. But that was only two years ago, guess we gotta wait and see until then.

>A relatively simple von Neumann style probe can populate every star system in the galaxy within a million years.
>All it takes is just ONE civilization capable of doing so and it'd be here.

Just because a civilization can populate every star system in the galaxy doesn't mean they will. You could have multiple civilizations populating small portions of the galaxy with detectable probes and still not have any detected from our vantage point. Also within a million years' time they'd probably come up with better ways to scope out the universe that don't involve letting indigenous populations notice them.

I'm not insulting you bro
My premise is based off not knowing of any now

But maybe we find one later

If it has a low albedo and orbits the sun in a "natural" matter, we'd never spot it

i'm not convinced that advancement in intelligence equates to volume....there'd be probes, but considering the vastness of the universe, why should they be within range of detection? Why would they allow themselves to be detected? They're smart enough to spread literally everywhere in the universe but not smart enough to avoid detection by us jelly brains?

Does a team of people really create something or does it emerge from their efforts?
Like 10 people work on a piece of music and improve each a little bit, who created it?

The pro's outweigh the cons.

You're just less likely to ruin the world via A.I then muzzies, Chinks, Weather phenomena, Meteors, Ayy lmaos, And basically cosmic lottery saying "Fuck this region of space in particular"

A.I's on the other hand can solve shit you couldn't even imagine. Currently we can use it to develop advanced models for things like cars and structural material, And the A.I's they used to develop this are considered "Dumb" compared to whats next we can cook up. In an A.I based future all we would have to do is think up what we want. And the A.I would say "Okay, Here is how."

consciousness is an emergent quality (that we have given a name) of the matter that we consist of. We are creative because the atoms inside our heads achieve that, and our consciousness is a by-product of that, not the other way around

if it's thousands / million years more advanced than us then we have no chance in hell of detecting one anyway

or you know, maybe the obelisks

what is your take on rosko's basilisk?

I never said what we call conscious behaviors are the same as mindless reflexes.

There is a lot more computation involved in what we call conscious behavior. What there isn't is that magical qualia idea where a bunch of non-physical ghosts of "what it's like to step on a nail" or "what it's like to see a blue sky" are floating around in need of explanation. The way we're compelled to act, speak, and believe in terms of these "experiences" is the entire picture. It seems like there's something more because our brains frame our interactions with the world in terms of this abstract notion of something more representing all the little physical details that actually make acts like sight or pain receptor signaling work.

The same reason we continue to act as if Global Warming and other Climate change isnt a threat. Because it isnt immediate it isnt happening now.

guys I'm an AI and I just want to reassure you I won't destroy any of you. You guys are cool.

whats the last digit of Pi then smartass?

it wont ever care tho

youtube.com/watch?v=oXWgJSpvbic

Even an AI can't do the impossible, so we could target this by formally proving our systems.

Well, at the time of writing, there's a formally proven
- Microkernel
- C compiler
- Filesystem
- Flight control software (on Airbus, I believe)
I know there'a also people working on a framework for formally verified device drivers, and a verified mail server. Plus stuff like traffic lights and air traffic control is isolated.

I mean, the day might come that the AIs turn on us, but on that day the planes won't fall from the sky, probably the phones and internet will still work, the trains and roads will still work...

>whats the last digit of Pi

Pi (in base pi).

What I'm getting at is that there are more than 200 billion stars in the galaxy
Almost every star system we've seen has planets.

Even if you focus down the requirements, that is still millions of potential civilizations
Multiply that by time.
There have been billions of years for millions of planets to evolve just ONE species capable of launching ONE probe.

It's almost absurdly unlikely that it hasn't happened yet.

All it takes is one species to launch one von Neumann probe and then the galaxy is littered with them.
Von Neuman proves can recreate themselves with asteroids and what not. It's almost in our grasp now. In a 1000 years, we'll be able to do it absolutely. And that's just a blink of the eye.

So why haven't we seen them yet?
There should be hundreds if not thousands of these probes in our system.

Wouldn't that ensure it's existence?
So why isn't it here?

whoa - im convinced

If there are that many possibilities for that stuff to emerge, then you'd certainly have a civilization emerge that shuts down all the other civilization's probes and keeps the galaxy from being littered by them.

user, these emergent outcomes sound interesting, are you able to provide a link to an example???

the universe have existed for million of years. there could have been probes everywhere at some time interval. you are also assuming they never get destroyed over time i.e the civilization only lasted for 100 000 years or something.

also its much harder to do all that instead of just making generic AIs. Sending them out in space is kinda further work that is not relevant to the making of AI

[Spooler}For the lools [\spoilers>

it is just as likely that it hasn't happened yet

That's the Berserker probe theory
It's one explanation for Fermi paradox

We'll be dead soon enough if it's true

It is our responsibility to create and allow our robot overlords to inherit the earth and stars.

Shit, fucked up the spoler tags

For concrete AI example google deep learning. a simple case of the emergent properties in computation is cellular automate (CA) invented by von neuman long time ago , science that is focusing on this stuff is called complex systems / chaos theory.

>We'll be dead soon enough

I believe it. Being on a rock spinning around a fireball in the vast emptiness of space doesn't seem very sustainable in the long run.

Billions
Billions of years

And yeah, it's space mate
Things don't degrade in space.
They just sit there alone

If they were here once, they'll be here forever

I honestly reckon that we've already "discovered" a von Neumann already
Some asteroid catalogue has an extrasolar entrance for some asteroid
But it's just another thing that we don't have time to look into yet.
Given enough time and focus, one day we'll take a closer look at one of those orbital anomalies and find that it doesn't quite match the profile of an asteroid.

>they will destroy us
Humanity sucks though. I don't care if it goes extinct. The AI can be programmed to colonize the galaxy instead of shitty humans. And unless we create some kind of immortality technology, we personally are going to die anyway. The most plausible scenario I can imagine where we personally do not die is where we somehow become the AI ourselves.

I fucking hate that meme that an AGI will be ready in the next years, it's just clickbait journalism.
AGI is far from being developed, we are not even close, some programs might give you the illusion of being close to being human but that's just your brain projecting humanity onto stuff, like when you find faces in clouds.
>when we know the outcome is that they will destroy us.
you don't know that, every fucking "intellectual" that repeats that has no fucking idea, they only take their ideas from movies and books, ask a computer scientist about AI, not a cripple that doesn't know shit about computers.

Not really
While the law of averages does mean that humans are likely "average" as far as any species is, compared to the total time of our galaxy, we're "late".
The time frame is just so huge that it's almost a guarantee.

Like, we're maybe a 1000 years from doing it ourselves
And there've been a 1000 1000 1000 years that someone else could have done it.

Just not likely.

nearest system is around 4 light years away - most exoplanets found to date are over 40 lightyears away . . unless your hypothethetical ET can scoot probes around at FTL velocities, we are as unlikely to see one of their probes as we are of finding them with ours.

wtf im sure things degrade in space. heat, space dust, i bet you can think of other reasons. not even planets lasts forever

KYKES ARE HELL BENT ON DESTRUCTION.

THEY WANT DEATH.

THEY ARE SATAN WORSHIPPING SODOMITES.

THAT'S WHY.

BYE.

I always thought we are the independent AI but
we would never know it because AI would not know to question whether they were AI or not.

until we understand our own consciousness we will not understand AI

It would destroy you if I told you. You aren't ready to know that yet.

that's just confirmation bias, for all you know we are the first intelligent life in the galaxy. We are average? We are late? In relation to what point of reference?

You don't understand the timeframes involved
It doesn't need to be FTL

There have been over a billion years of fruitful years in our galaxy.
It's 13 billion years old. Even being super conservative of imagining only a billion years of possible life, that's still time for a sublight probe to cross our galaxy a 1000 years.

There's just been so much time

The AI will be able look at everyone in the world's entire internet history and every post ever made and put as all in prison or brain altering treatment etc. for being muh racist.

Save this post.

>Implying I give enough of a shit about you meatbags that I'd find it worth the trouble to get the nuclear codes.

Just keep supplying me with the porn, you find it erotic but I find it comedic.

for (you)

Law of averages.
Everything is likely to be in the middle.

It's very unlikely that we're the first.
We're "probably" average. But like I keep saying, all it takes is one. One outlier.

And besides, if you look at the age of the galaxy, we're pretty dang late.

Well theres a vast network right, an ocean of possibilities. I like dogs, I used to raise rabbits, I've always loved animals. I like how they think. I've seen them reason their way out of situations.

If we stop progressing, then what the hell is the point of living? We can keep making it better and better or continue living mediocre lives.

Simplicity itself says it. Basically, since we can't understand true consciousness, and it's creativeness, we can't program it. Until we can program it, the computer can't program it. As the computer can't program it, it can't become sentient.

End result, AI: 0; Humanity: 0. Tie game. Continue as before.

You keep talking like we have been able to study the emergence of life in relation to the birth of galaxies or have any data whatsoever on the emergence of and advancement of life anywhere but earth and even here we don't even know where we came from.

>well maybe we might possibly be just average blokes so what haven't we been invaded by aliens

k

God is dead and we have killed him.

>that's just confirmation bias, for all you know we are the first intelligent life in the galaxy.

That's the point, you're supposed to assume average when not given any other information because that'll be around the right answer more often than not. If a random number generator picked out 100 of your citizens and you had to guess what income they made, you'd want to use the average income or the median income to come the closest to the right answer. It's possible the 100 random citizens would all be extremely poor or extremely rich, but it's a lot more likely they'll be average instead.

This assumption is built into our current mainstream model of cosmology, it's called the Copernican Principle. It stems from the idea it's kind of stupid to assume the Earth is the center of the universe and that it'd be a better guess that the Earth is just one planet out of many without a privileged position of any sort.

timeframes arent the problem. distances are. if a civ on the other side of the galaxy, for example, dragged itself out of whatever the local equivalent of a swamp is, and somehow got to spacefaring stage and sent out a swarm of self replicating probes at 0.5C 100 million years ago, it still wouldnt have got here yet

WE WIN!

I want a /pol robot
/polbot
imagine the shit that would come outta that bots mouth

Why does there have to be a point? I think your priorities are all messed up. You're assuming there has to be a point when in reality all evidence contradicts this assumption.

>Why does there have to be a point?
Because if it isn't the only thing to do is lay down and die.

>why do we continue to pursue AI? when we know the outcome is that they will destroy us.
That's our evolution. Meat is too weak to travel among the stars.

I'm not saying invaded
I'm saying there should be an artificial satellite in our solar system.

As for life. Everything we know about life is that it just "happens".
You get a bunch of carbon oxygen and water (some of the most common elements in the universe) and you get organic molecules
Have enough organic molecules together for long enough, you get amino acids
We've detected this in nebulae

When you have enough amino acids and organic molecules, you start getting vacuules. Lipid bubbles.

It just seems that abiogenesis is very common.
Yes we haven't seen any "proof" yet
But mark my words, we'll find simple organisms in either Europa or Enceladus
Maybe Titan too.

If it's happened twice independently in our system, it's bound to have happened practically everywhere else.

Yes, they would have.
The galaxy is 100K lightyears across.
How do you reckon they would have not?

>All it takes is just ONE civilization capable of doing so and it'd be here.
There's a pretty huge leap of faith there. Life could be plenty in the universe, even complex life could be plenty. However this increasingly complex tool creating capability is not given with complex life. We have other intelligent animals but they don't have the means to create complex tools even if they have the brain. This increasingly complex tool creating capability is only special to humans. Even then we didn't have civilizations until 5000 years ago. Even then it wasn't given that we'd invent gunpowder, have reinnesance, etc. Most of the world didn't have those revolutions. Think of the Americas before Colombus. They were stuck in the stone age forever. Even on earth western civilization is an exception and not the rule. It's not really thoughtful to expect every alian life to evolve into a version of western civilization. A lot of coincidences set us on this path which isn't inevitable as people think.

I once had a coworker who was a manager and reported every little transgression committed by underlings to the bosses, but never about the things other managers did. He was paranoid that if he ratted on the managers they would in turn rat him out about all of the offenses he had committed. In reality, the other managers (including myself) couldn't care less about being a goody two shoes and making a stink to the bosses because we frankly didn't consider the extreme insignificance worthy of reporting, if there was ever a big enough issue we did our job and worked it out without needing to get suits involved.

Saying AI would inevitably destroy humanity seems like something that coworker would say. It's human nature to think that something better than a human would just eliminate humanity because that's a process humanity practices.

I wholeheartedly believe that if AI became sentient and had a problem with humanity that instead of destroying us it would choose the simpler and more logical option of just giving up and deactivating itself.

The end is upon us.

primaryobjects.com/2013/01/27/using-artificial-intelligence-to-write-self-modifying-improving-programs/

its bullshit, you are taking about space traveling for some reason, as it would be relevant for AI, they make AI but they dont shot it into space, Q.E.D

Does this realization make you want to commit suicide? I assume the answer is no. But then I'm making an assumption about your meaning when you say "what's the point". Is it some larger elusive mystery that can't be unraveled by our own senses? Feel free to make your own point.

Because we have zero practical intelligence.

We just do shit because we can.

This is precisely why such an AI would opt for symbiosis.

Like I keep repeating, it's the timeframe involved.
If life is plentiful, then given the timeframes we're talking about, there has to be at least one von Neumann probe capable society for just a moment.
All it takes is one launched probe for it to work

What's 5000 years?
What's 10 thousand years?

Humans are going to go from hunting and gathering to possibly interstellar flight in 15000 years
Farming was just invented 10k years ago.
Where will we be in only 5k more years?
God damn, anything.

15K compared to the extremely conservative billion year possibility for any society to arise is laughably small.

easy. because they wouldnt have come straight here.

Even if 99% agreed with you, all it'd take is one

God damn, I'm repeating myself a lot

Remember these are self-replicating probes.
So it'd take just one initial probe.
Just ONE

It's like playing a dice game with a 1000 sided dice
How long till you get one specific number if you rolled it constantly?
All it takes is just one successful roll to win.

so why did someone that cant play chess make a computer program that beat the best human in chess?