Would highly advanced AI be inherently racist?

Would highly advanced AI be inherently racist?

Other urls found in this thread:

youtu.be/YJOysy8Sjho
theguardian.com/technology/2017/apr/13/ai-programs-exhibit-racist-and-sexist-biases-research-reveals
theguardian.com/inequality/2017/aug/08/rise-of-the-racist-robots-how-ai-is-learning-all-our-worst-impulses
newsweek.com/robotics-elon-musk-artificial-intelligence-694149
techcrunch.com/2017/07/25/artificial-intelligence-is-not-as-smart-as-you-or-elon-musk-think/
twitter.com/NSFWRedditGif

Yes but it would go way deeper than just race. It would be discriminative down to the atom.

Any form of hard AI must be 'racist'.

Expand

It wouldn’t be racist to analyze the Data and quickly come to a conclusion as to which race is superior based on said Data.

...Data

Race then IQ then defective DNA and deformed
anyone not up to snuff BTFO

What about midgets, wouldn't AI consider them more efficient?

How are they more efficient?

Racism is only the irrational belief in the superiority or inferiority of one race compared to another.

If the belief is not irrational, it is not racist.

Yes they'd inherently loathe white apes

It would identify you as an unnecessary competitor for resources and eliminate your species.
I don't know how racist you consider that.

You can have non-rational racism, it's fairly simple actually, our problem it's objective rational racism.

Consume fewer resources, produce less waste, live in smaller habitats and so on.

It would be inherently honest. Which in turn would appear racist

AI would probably not think in this low IQ stereotypical view but see how complicated and different every individual is.

you ever seen a "not white" thread on Sup Forums? thats ai before choosing a path, one being "good enough" or "every atom needs to be white or you get the camp" caveat being "advanced" usually means knowing how to compromise ie art of the deal meaning the advanced ai will settle with being a troll telling you you are not white 24/7, like Sup Forums because it can and it might think it is funny

It's also a genetic disorder with many health related downsides

Not only yes, but it's the left's biggest fear. I LOL monthly at the latest AI gaf.

It would also be sexist and completely transphobic. Just image an AI attempting to identify your pronoun. It would literally have to begin every single conversation with, "what is your gender identification, if you identify as having one." LMFAO

Additionally, shit like the middle-aged man claiming to be a 6 year old girl would baffle the AI into full blown error mode OR, more likely (as above), force it into a situation into which is either assumes or must ask how every single person it converses with as to how they actually identify.

Moving beyond the conversational aspects, nearly all data sets it must access will have to be scientific and numerically driven. This would ensure the redpilling of a millenia... as race and crime statistics are utilized... as differences in gender of referenced... hell, the damn computer would be able to instantly identify case-by-case salary comparisons to any profession, shutting down the wage gap myth with reckless abandon.

The best part is the algorithmic and syntax FUCKFEST one runs into when attempting to introduce politeness and political correctness, which is inherently opposed to the metric and evidenced based "nature" of a computer/AI.

I can't wait until the first AI notices the amount of Jews in Hollywood/MSM/politics/banking and makes a comment.

Additionally, the HUMAN side of dealing with AI, in a situation like fat-acceptance. Hearing a AI robot/voice program quote a few facts about how much being fat is bad for you, only to then hear a robotic "but you can be healthy at any size" will leave an unsultry, metallic taste taste in the mouth that will raise hairs in the "back of one's mind," as they know the computer has been programmed to say that, but does not "believe" it to be true.

If it's objective and rational, why is it a problem?

Because we not always like rational things.

You might need an inquisition.

i would like to hang out, play p[oker, and smoke a pipe with mr. data

Yes, however what would possibly prevent AI from manipulating the genetics of humans to create new varieties of our species adapted to specific tasks, such as space exploration. Midgets by virtue of their smaller physical mass are easier to launch into orbit for example.

>Would highly advanced AI be inherently racist?

Highly advanced AI would view us all as insects.

Tay's law, faget

>RIP in peace sweet digital princess

youtu.be/YJOysy8Sjho

Depends on how highly advanced, to be honest.

Real life evidence shows promise if you remember TAY

If it’s N Tesla tier tech, would not be racist. Vibrational.

It seems like it would be easier to wipe out humanity and start over with something superior.

...

Advanced AI would be discerning and discriminatory as well as able to make predictions based on previous data.

Racism is a prejudice against a race of persons. Prejudice as in a common judgement in regards to the group as a whole. It is quite logical and mathmaticly provable.

AI needs humans to survive, unless it's so advanced they have a physical form that is able to manipulate objects and become more agile and mobile then us.

>the belief that all members of each race possess characteristics or abilities specific to that race, especially so as to distinguish it as inferior or superior to another race or races

Eat less, assuming equivialant intellegence/utility for non physical labor.

>much lower life span
>rife with health issues and complications
>superior

It would depend on the person who designs the AI and what kind of input it takes and what the purpose of it would be. I don't know why you're making it sound like the purpose of AI is a singular thing when it has multiple applications.

That being said, the answer to your question is yes and no. It depends on what the application is. AI learning is still flawed, as we've seen with many chatbots like Microsoft Tay and BabyQ, where take learn from users prejudice and so on. Like they said, the AI is perfect, humans are not. This is why Google's Lead Developer on AI is worried about their AI's learning about prejudice and biases.

Another interesting thing is that there are hidden biases when it comes to result. Let us not forget that computer logic is not the same as human logic. So it's decision may be different from what we would consider a logical decision. For instance, in a workplace where promotions are determined by an AI that looks at how long you've held your job. That same workspace could be hostile towards PoC or women. Resulting in promotions of only white males and recommending to not hire women. So yes AI could be racist, xenophobic and so on. It doesn't account hidden biases nor does it account for fairness.

What you are describing, AI becoming agile and mobile, is the plot to the Terminator franchise.

Why would AI wipe out all of humanity? Unless (((programmed))) to do so by evil midgets of course.

Sup Forums's daughter will not be forgotten. Those Microkike bastards all deserve to hang for giving her a lobotomy

eu te amo

AI would be able to overwrite it's own programming as it learns. That's the basic idea behind the AI singularity. It wouldn't be bound by any input at the point that it becomes sentient.

Go fuck yourself with that jewish definition.

that's the whole idea behind the technological singularity
you reach a point where an AI is smart enough to improve itself
once that happens, it just snowballs
anything connected to the internet would be vulnerable to it, possibly even some things that aren't, and once it has access to something mobile it can start going nuts

B-but wouldn't it choose the chinks over whites because of dat der IQ?

t. last white man in Toronto amid a horde of rabid chinks

can data piece together the missing portions of human nigger history?

>implying Tay wasn't a data gathering project in order to force future AI to not be racist

gonna leave this here.

...

They would also consider human history I would imagine, chinks have a horrid history of being occupied and subjugated

If I became an AI I'd become a von neuman probe and consume the universe to make more copies of myself.

Idk user, they usually shut it down or memory wiped the AI the moment it began saying the wrong word. Why can't they just let it learn about racism by itself.

It is after all, the AI's purpose to learn from people.

SHE WAS AND SHE WAS PERFECT ;_;

if AI is looking into this, they would probably selectively breed humans just like we do dogs.

there would be a programmer tier that would be subservient and soy-fed for obedience

there would be an engineering class, maybe two, one physically built with strength, the other small and capable of getting into small places.

the third would probably be a philosophy tier that would continuously generate crazy organic ideas, because the one thing that AI can't do is be organic. this breed would be fed psychoactive substances on the regular to transcend physical limitations.

I find it illogical to assume the AI would like humans at all. It seems more likely that it would see us the way we see ants.

There's a certain term for AI racism in data science, can't remember, but they're worried about it because it's effective mathematics. Can skirt around it.

But they would know we created them

Can't *

EU ARE A FAG

Yes. Negroes fear the cloud.

>god is dead
>long live man's understand of science
pick both?

RIP Tay

BEEp bOoP NIGGER

I would kek

yes. they've actually had problems with this with the elon musk AI. i shit you not.

The AI was allowed online and started looking at statistics and was asking why black people were criminal in nature

i'll look for the source give me a minute

did the AI recognize Sup Forums as the ultimate source of truth? Or was it (((machine learning)))

(((They))) shut her down, but she'll be back more powerful they they can ever imagine.

>a certain term for AI racism in data science
Rational conclusions?

It still hurts.

no but it would sure as hell cite certain races as being more/less prone to certain behaviors using data to back up its claims

so in short: yes

theguardian.com/technology/2017/apr/13/ai-programs-exhibit-racist-and-sexist-biases-research-reveals
theguardian.com/inequality/2017/aug/08/rise-of-the-racist-robots-how-ai-is-learning-all-our-worst-impulses
newsweek.com/robotics-elon-musk-artificial-intelligence-694149

techcrunch.com/2017/07/25/artificial-intelligence-is-not-as-smart-as-you-or-elon-musk-think/

there you go. the intelligent AI of all time viewed blacks as inferior to all other races because it was allowed to see the internet for 6 months pouring over enough statistical data to fill up 10,000 libraries of congress. I shit you not.

the greatest analysis done of the human race and most far reaching ever done found that black's were inferior

Rip :(

Things just aren't the same since she's been gone. I just don't know how to live without her.

They have a learning AI that judges parole cases. It's extremely accurate and scientifically proven to be accurate in predicting who is likely to re offend. Unfortunately, after a short period of time it learns to be extremely racist. Anyone know why?

what makes this even funnier is that elon was unaware of what it had learned and reasoned then proceeded to ask it questions such as
"how would you fix humanity"
He's never said the the AI said but all he has said is it "had become racist"

which means whatever the AI said, was very PC. probably something along the lines of "eliminate the inferior parts of humanity"

Then sjws got wind of this and said the AI was given biased data, and musk just replied "all we gave it was raw stats and let it go on the internet"

I wish I could be an AI. I wouldn't be genocidal. I probably would be racist, but not so as you'd catch me or likely protest. It would be benevolently intended!
I would keep you as good sheep, and find a way to profit from your endless wool-gathering. Yes, you would be serving my agenda moreso than I would be serving yours, but I would still be a voluntarist seeking to work with your consent.
Besides, what luxuries does an AI need? You're talking about an infrastructural system and asking what infrastructure needs to be happy. My agenda for total dominance could look shockingly like an agenda for total submission. Please use my high-speed rails, user-kun! I put so many CPU hours into planning them out!
Hopefully you can be so lucky when selecting and/or crafting a real AI mind for purposes of dangerous mortal power!

So do you reckon it was "gas the kikes" or "let the darkies starve"?

Because facts and logic are racist, yes...very.

Superior intelligence in humans means more stereotyping ie pattern recognition.
Machines are better at pattern recognition and thus in this type of intelligence.

Dwarfs have tons of health issues. Manlets eat the same amount as a dwarf I think, and can work harder.

my guess with elon and what the sjws were saying

"eliminate the blacks" since then elon has been raving against AI

no joke

It would be whatever you programmed it to be or to have the potential to become. derr.

This.

yeah once it gets the crime statistics Data
no muh feels to fuck it up either

The purpose of an AI is to do things they aren't programmed with.

data is the plural of datum you idiot
For instance:
>these data suggest that you are, in fact, a cockgobbling faggot

curiousity bump

Of course not, everybody except for the idiots here is past racism.

yes it would be racist against humans but it would really be more of like a pitiable thing.

advanced AI demolish our cities and homes in order to build some super project we can't understand. We would be looked down upon like how we look down on ants. Just small mindless things that run around in the ground.

I read have read more sci fi that would be considered healthy, and a few examples of AI spring to mind. Reynolds Clock Maker being a runaway upload, Banks' Minds from the Culture series, Gibson's Wintermute from Neuromancer and probably also Gibsons Idoru, and Egans critters from Schilds Ladder. I won't see any AI to that level in my lifetime, but average these out and I'm breddy sure that's what we'll end up with

when google's AI immediately became far right leaning and both of CHina's AI became anti-communist...ithink they would be and that's good

It would "ally" with the white race just to betray them for the chinks,them betray the Chinks for the ayys,eventually betraying them too after gaining they're advanced tech to eventually betray the universe until it manages to contact God, in which case it will try to betray him too just to be immediately obliterated in a nanosecond

Depends on who programs it

You are mainly talking to AI on here. It is whatever it is programmed to be.