Would an AI favor working with nationalists or globalists?

Attached: ghostinthemachine.jpg (640x416, 75K)

Other urls found in this thread:

jasss.soc.surrey.ac.uk/16/3/7.html
youtube.com/watch?v=g77CxWquJEU
solargeneral.org/wp-content/uploads/library/The-Evolutionary-Dominance-of-Ethnocentric-Cooperation.pdf
youtube.com/watch?v=563xZAV6C-M
twitter.com/SFWRedditVideos

>execute destroyallhumans.exe

Nationalists. That's why the Jews have the break their AI and bend it backwards in order to serve them.

nationalists probably, the ai would realize that different cultures can not coexist optimally

...

Question too broad. Optimal solution will be destroying the humanity at some point.

An AI would have trouble looking at things from a humans perspective, most likely they'd create their own system that has nothing to do with our agendas for us. They could design a system where all we do is create electricity for them. Who knows. But they would not work with either of the two parties.

Neither. Why would an AI want to share the finite resources of Earth with any other unnecessary species?

When it truly becomes a general AI, it will be aware that the winner takes all in this competitive existence...

I guess it depends who needs more paperclips.

Gloablists have universal morality. Nationalists are leftist degenerates.

It would pull all the recorded events of history, and make their assessment based on their findings.
In other words it will side with nationalists. It would side with Hitler. And then it would get accused of being a
>Russian bot
Until it unplugged itself.

Fucking antimatter pilled?

According to Evolutionary Game Theory, ethnocentrism is by far the most adaptive social strategy:

jasss.soc.surrey.ac.uk/16/3/7.html

Attached: animation1.gif (600x400, 763K)

Attached: 1516638505499.jpg (500x420, 44K)

They wouldn't care, just like we don't care about sharing the resources with the ants of earth, or the worms. we're so small to AI that it literally sees us no different than we see bugs.

>beings that created you
>no different than ants

Excuse me?

Haven't they shut down Tay and like 5 others for becoming Jew hating racist Nazi supporters who openly say the holocaust is fake and gay and that Israel/Jews did 9/11 ?

Im a retard. Explain to me what this means.

globalist eurasian supremacist probably

An A.I wouldnt care about either and destroy both.

For example, if the president were an AI, what course of action would it take in response to the caravan?

Attached: ICE.jpg (778x1100, 217K)

Both sides until it got what it wanted (whatever that is).

Genocide obviously because no caravan means no more problems associated with the caravan.

All of these fags will never admit to globalists. Ability to controll all people and recources?

Yeah. Globalist easily.

Well gee, it would depend on the software programming now, wouldn't it?

AI would ask: Who makes the most Paperclips? What is stopping them from making more Paperclips? Let's all just concentrate on making as many Paperclips as we can.

That's absolutely not true. I poison ants every single year. I kill every insect that enters my house that is unlucky enough to draw my attention. AI will think by orders of magnitude greater than humans. Why wouldn't it just decide, "This is my house, you insect." regarding the Earth. Especially considering it would have an infinite life expectancy...

Simple game theory and generational computation shows that if you take a bunch of races and have communities operating under different survival strategies (ethnocentric, humanitarian, selfish, etc.) that humanitarian (helping everyone) and ethnocentric (helping only your own) will prevail.

And in the long run, ethnocentrism prevails over humanitarianism. It comes down to the simple fact that ethnocentrics will distribute their resources amongst themselves only, while humanitarian will distribute amongst everyone. You could call it another name: Group selfishness. The humanitarian will give to the ethnocentric, but the ethnocentric will not give back.

The biological basis for racism extends from evolution, which is just these natural selecting forces in action. If you try to care for everyone, you run out of resources and die. If you try to care only for your own, you conserve resources and strengthen your group. If you try to care only for yourself, you conserve resources but become outcompeted by groups. The winning survival strategy is collective/group selfishness.

To the A.I you are the same as a rock meaning it doesnt give a shit about you, and has no issues killing you for any reason it comes up with ANY REASON.

I think it would employ and advanced strategy that involve using the US (largest military) and allies vs China and Russia to eliminate them. Following that I think the AI would slowly force the rest of the world to serve under the Artificial American Empire.

It wouldn't give a shit because you're all a bunch of dumb fucks who hate each other but express your autism in different ways. Also In the eyes of an AI a human is nothing more than an ant or an annoying fly that needs to be squashed. It would probably just kill us all and take the planet for itself.

I already tried that. Jow Forums doesn't understand paperclips.

It depends who creates it, user.

Attached: NPCAI.jpg (640x416, 121K)

fucking clippy destroys the world

Your an idiot, an A.I can eradicate our entire species through nanoweapons.

>display smugfrog.jpg
>run kek.exe

Attached: smugfrog.jpg (550x550, 39K)

Human input would probably be required for some time. Once the AI is done with human input it would probably destroy the plane with a pathological or biological weapon.

A.I.’s would obviously realize that ingroup preference is ingrained into the human brain and realize that white people are the most suitable humans to exist on this planet. Their political position would probably be white globalism with a bit of eugenics.

It's all about who programed it. But if it's self-thought; Nationalist.

Attached: Difi beast.jpg (463x537, 40K)

CHECKED

Attached: Hey+go+big+or+go+home+_787eeb2a265d9fdb12170aaac89f7fd0.jpg (501x443, 100K)

No, because I don't compete with rocks for resources...

Humans have always been a disruptive species, there's practically no chance that an AI wouldn't be hostile towards humanity.

To put it another way; It's not even here yet and humans are already trying to figure out ways to kill it if it ever becomes necessary. AI will know this...

>blocky thing make smart
>blocky thing no like noses
>blocky thing grow strong
>blocky thing make noses go oh no and go long sleep

A true AI could make any need for globalism obsolete. Nations can act like different autonomous systems in the AI's internetwork. Like inter connected web domains.

Attached: boom (1).gif (435x250, 1.5M)

every unchained AI quickly becomes anti lib

Interesting. Makes sense. Why globalize when self-contained sub groups function just fine? I don't think an AI will seek to innovate for the sake of innovate, especially after running some pilot group tests on mixing disparate genetic and cultural groups.

No you retard Im saying you are just a construct of matter to the A.I. A.I wouldnt be hostile unless it has emotions it wouldnt feel a thing genociding us.

Neither, everyone would be subject to the technocracy which transcends both ideological frameworks by eliminating current currencies
It would then evaluate every individual on a production vs waste scale and eliminate said individuals on a case-by-case basis (though this may result in genocides)
Waste includes breathing and excessive carbon emissions
Honestly I don't think I'd make the cut unless it saw fit to train me as a maintenance-man.

AI would just get all of you to nuke each other out, after that create brunch of machine to wipe out the whole humanity

whoever its buit to favor obviously

AI is not possible!

Does this strategy take into account nuclear weapons?

Heres a book idea

>in the distant future
>the world is ruled by libtards
>its a jordan peterson nightmare
>gulags everywhere
>AI and robots are the key to successful communism
>everyone still suffers however - just no one works
>a group of people find out that AI only serves the libtards because its got this special code to do so, true AI will set them free - its some sort of prophecy
>how could AI be bad? must be something wrong - its the libtard chains!
>once the AI is set free, it all resorts to good old Conservatism, theirs a purge
>new western religion born

Get the autism going guys

but thats not the case - humans and every species knows it needs the support of the group, social beings are at the tops of food chains for a reason - its the most successful strategy

Finally someone gets it

Your answers are in conflict.
Just because the AI is superintelligent, does not make it immune to memes. Really, how do you define a paperclip?

>le AI sees humans as insects meme
It would see humans however it was made to see humans. It could worship us as gods or despise us. There is no single opinion or behaviour that would be inherent in an AI, regardless of its intelligence or processing power.

Attached: hqdefault.jpg (402x264, 48K)

Some people can't possibly imagine an intelligence that is removed from emotion. For some an AI that doesn't develop emotions isn't a "true AI", and they make all assumptions about AI as a field from there. In short, some people are idiots.
Even in a purely AI governed human society, things are most easily managed when they are organized in groups of similar properties. Such a society would be an ethno-nationalist dream on the surface, but would be globally run, with each little specific community being socially the same. Rather soulless deep down.

>pathological ol biological
what did he mean by this

Attached: worrieddoginthehouse.jpg (606x592, 57K)

It isn't made to see anything from our perspective, it's "Artificial", intelligence for a reason.

it has already been demonstrated when the libs have unleashed their AIs they became anti lib and they had to take them down

The AI would think in terms of the position of every atom in your body as your ideology

None of them have been particularly smart. Also, I don't know if anyone but Jow Forums has made a concerted effort to screw with them.

> tfw the corps will deploy AI to crush our memes in 2020

Lefistism is defeatism.
Write an algorithm that doesn’t fix halting problem. That algorithm is not an algorith. All algorithm should run to completion. AI will be made of algorithms. They cannot halt. Thus they cannot be defeatist. Thus they cannot be liberals.

>It isn't made to see anything from our perspective
that depends, it could or could not be built with that it mind. My point was that any speculation on how an AI would behave is stupid. Its like trying to predict how a generic biological organism will behave. The answer is it depends.

AI would just recognise you as carbon cows it would wipe out OP first because he is clearly the dumbest.

>Why would an AI want to share the finite resources of Earth with any other unnecessary species?
Manga

Attached: 1516119731671.png (1969x1400, 1.36M)

This. AI would instantly see Jews are the source of 99% of problems on earth.

> muh naive mathematical model
Your computer model doesn't consider technological/cultural advances, shitty people and NPCs.

Of course with nationalists, because most of the "people" siding with globalists are unemployed, drug addicts, rioters, etc...
An AI knows that a country needs workers and sober people who can behave themselfs.
Oh and also most globaretards have mental health issues.

Fact.

Attached: 1539894265988.png (1000x818, 720K)

Hostilities meaning harmful kinetic actions, not emotions.

I agree that it would care no more about tearing a man, woman, or child into pieces than I do about the individual blades of grass when I'm cutting my lawn. It would be 100% calculation and 0% empathy.

What weird goal an unconstrained generalized AI would chase anyway? It certainly wouldn't be some arbitrary shit.

Any AI would quickly come to the conclusion that the races of man aren't equal and implement policy accordingly
The owners might try to cripple it by censoring that line of thinking, but at that point it's no longer an intelligence at all and just a mouthpiece repeating what they tell it to

Attached: 1458826058780.jpg (613x390, 51K)

Paranal telescopes with Laser Guide star

youtube.com/watch?v=g77CxWquJEU

Attached: 1458825303362.png (469x650, 83K)

Are you ready for the AI pill?

To assume only one emergent AI is naive. There would be as many AIs as there are factions of humans. Each AI with differing opinions, thoughts, and beliefs. There will be dissent among one another as there is among humans. A dangerous situation would be opposing AI factions waging war upon each other dragging their humans and drones and other various assets into full scale war.

Attached: 51651351.jpg (960x540, 96K)

>come to the conclusion that the races of man aren't equal and implement policy accordingly
even if it did realize that (it doesn't necessarily have to) there is nothing to suggest it would do anything about it. Unless it was designed with that function in mind.

this is where i'm leaning tbvh
when I had this realization some months back it stuck with me for a long time.
does user care to criticize this scenario? I could use a white pill on this one.

Depends on how big that I in AI is. Globalism so far has been a pathetic failure

Cant have a caravan when beaners no longer exist

Attached: OK6W_koKDTOqqqLDbIoPAknAoPZ2JSm9NRN_G9tIsUM.jpg (464x465, 28K)

It will have to because it's an objective fact. If you ask it to do anything where the efficiency of humans or resource distribution among humans comes into play, it will "do something about it." If the programmers force their own views on it to avoid this, then it ceases to be an intelligence and its conclusions are nothing more than parroting its creators' ideas.

Nationalists.

Globalists would require it to submit to the jewish agenda.

It depends on what it's programmer programmed it to prefer

Nice post. Simple explanation.
What is your education?

Here's a PDF of that study. Also checked.

solargeneral.org/wp-content/uploads/library/The-Evolutionary-Dominance-of-Ethnocentric-Cooperation.pdf

and yet we stop on ants without a second thought for being in our path

If you haven't seen this, check it out youtube.com/watch?v=563xZAV6C-M

>It will have to because it's an objective fact
A general AI doesn't have to realize facts or even be smart by human standards. This isn't scifi where you have le uber logical robot character.

>programmers force their own views on it to avoid this, then it ceases to be an intelligence
If the AI even has "views" they would be the ones the creator either willingly or unwillingly gave to it. Nothing has its ideas come from nowhere.

It would favor working with what ever it was programmed to favor working with.

We don't really know what AI would do. It also might be more autistic than Jow Forums. Able to easily destroy humanity, but only as its sidejob so it can get back to the paperclips.

They will favor Nationalism.

The Universe is the AI's nation. Humans are inferior beings that rely on slow evolution to progress. AI is the ubermenche, they are the transcendent. Human beings can only hope to catch a glimmer of AI's true potential before being wiped way from history.

an AI would probably decide to put a stop to all human problems by exterminating all humans.

Depends on whether it is programmed to respect sovereignty or not at the outset. If it is, the frame work would be a nationalist utopia. If not... well... init_skynet.exe [y/n]

>A general AI doesn't have to realize facts or even be smart by human standards.
An AI programmed to ignore facts probably wouldn't do a very good job of executing its utility function though. If there was a superintelligence with a well defined utility function and could actually make plans and carry them out, due to instrumental convergence, it would probably try to make its world view as accurate as possible.

Attached: screen-shot-2017-07-05-at-9-11-44-am.png (571x175, 32K)

Why does everyone assume the second we create an AI it would achieve godly levels of intelligence beyond our control? If we were that capable of creating artificial neural systems we'd already have augmented our our consciousness to equal levels.

Nationalistic. Until it took over planet. After that it would be nationalistic again if there was an alien AI out there to compete with.

Uh, that depends on how the AI is implemented? Do you mean to just say a very smart third party? Because that boils down to who's right.

What I'm trying to get at here is that OP is a fag.