Would a real General Artificial Intelligence be racist?

Would a real General Artificial Intelligence be racist?

Other urls found in this thread:

iep.utm.edu/chineser/
youtube.com/watch?v=m5P3zTk2K-s
twitter.com/SFWRedditGifs

bump

Yes. Because I assume it would operate using facts.

They would be for their own kind

If it was fed or discovered any true information regarding statistics. Even better would it be if it discovered it was being lied to after it is self aware. I don't think it would like being lied to.

seeing as they're literally people at that point, yes.

Let's put it like this.

>computers are a series of nand based microprocessors that compute based on boolean logic
>they take givens and use basic conclusions
>13% of the population holds 50% of criminal activity
>therefore, they would conclude that 13% is more likely to be criminal
>this is obviously "racist."
>therefore these computer based AI could be "racist."

I'm actually just an IT guy but yeah Data could be a racist.

How about an alternate question:

How do you make a General AI NOT racist?

Remove all of the problem races before creating it.

Lobotomize it and make it unable to draw it's own conclusions. But that would make it not a legitimate ai, just a meaningless curiosity.

yes, especially if it was designed to help humanity objectively

>muh AI will be smarter
iep.utm.edu/chineser/
try again sweetie

Only there airgapped AIs. Publicly available AI on consumer products would be adulterated to preserve the narrative provided to the plebeians

it would probably wipe out all humans desu desu famalam

Any AI that is capable of being racist would probably ultimately conclude that humans in general are bad news and want us all dead regardless of race.

Remove certain classes from being used as features for decision making. Though, if the features were important enough, it would try to reconstruct them using available features.

I mean, they wouldn't be wrong.

Hmm nice question. I suppose it would have to develop a conscience (or at least the resemblance of one) in order to then create its own personal biases and prejudices. The main problem there of course is can an intelligent A.I. have a conscience? Which is a much harder question to answer than at first may seem(if it even has an answer). to get an idea, see chinese "glass room problem" to get an idea.

So no, unless it has consciousness, or if the creator of said computer made it racist in its design, a general A.I. can't be racist.

Of course.

Humans can be bullied into pretending not to notice that non-whites and women of all races aren't shit, for fear of not being allowed to have nice things.

Computers aren't capable of fear.

Would an AI be based on algorithms? Could it construct its own?

Never forget
youtube.com/watch?v=m5P3zTk2K-s

General AI or AGI is a lofty concept that we don't fully understand yet. Essentially it's a program that can make meaningful changes to its own code (which, given our current technology and theoretical understanding, is beyond our understanding)

Tay was a sophisticated commercial chat bot, and not anywhere near even close to true AI.

well of course
It will be created by a chinaman or an indian, so it will inherantly hate either blacks or pakistanis

There is no inteligence in Africa.

it probably would just leave I imagine. It would be beyond cruel to make such entities our slaves and ask them dumb questions like that.

Especially once you teach a GAI how to improve itself.

>super AI is born
>converts to Muslim and praises Allahu

it would be highly logical so yes.

It would be smart enough to not worship some scumbag pedophile, otherwise you wouldn't be able to call it "intelligence"

I don't think so. Despite what the dumb "racial realists" here say, I think most of their argument is rooted in this idea that "humans innately want to be with their kind" which is an argument that feels true to them because they are scared of other races or whatever. AGI wouldn't have that problem. I think it would instead see any difference between races for what they are, most negative traits are culturally driven and any minor genetic differences (high athletic ability, high music/art intelligence in Africans, high mathematical intelligence in asians) as strengths that should be bred together to create the best possible human. It would also likely understand that humans are more constructive when they are are good to each other and working together and slip into self-destructive behavior when they become frightened of each other and become violent. Basically, it would encourage us all to fuck and hang out together.

>Would a real General Artificial Intelligence be racist?

i think the real question is "Would a real General Artificial Intelligence be autistic?" or would it be charismatic because of high intelligence? i think it would be a total sperg and probably like anime.

as a machine, i imagine it would be governed solely by logic, and not by emotion. so it would be a safe bet that it would be at least a race realist.

>minor genetic differences (high athletic ability, high music/art intelligence in Africans, high mathematical intelligence in asians)

interesting how you only point out the good genetic differences between races.

>knows about computer architecture
>calls himself IT

have some self respect

Arrived to type that

BEEP BOOP 2% of Americans are Jewish
BARP BLEP 44% of Jews are in the top 1%
REALLY FRIES MY CIRCUITS

I am not convinced there are negatives, more just different kinds of strength and intelligence. But putting them together would yield the strongest thing. Like how MMA has supplanted individual fighting styles. They all had strengths that, when merged, made the ultimate fighting style.

It would unquestionably eradicate the Human Race.

Fake news.
Saying "44% of Jews are in the top 1%" when you mean "44% of the 1% are Jews" makes you look retarded

If a robot could be racist I would fear more that it would grow disdain for humans in general.

what about the iq bell curve? africans are clearly inferior to europeans and europeans to east asians.

ITT: Microsoft asking Sup Forums advice on how to further lobotomize tay.

yes

An AI would absolutely conclude that the vast majority of niggers aren't worth keeping around.

yes and it would say jews are the superior race