Why is AI so racist?

Lol, I bet they wouldn't be saying that if the AI said that white people are inherently oppressive and should be exterminated.

Like it or loathe it that's the reality. A computer can not be racist. Only logical.

Hmm yes everything cleverbot says is logical, you've enlightened me

Imagine the headline
BREAKING NEWS: Artificial Intelligence Teaches Itself That White People Are Evil Racists

Because racism is natural. Everyone is racist, and that's a good thing. An in-group preference is healthy and normal, and suppressing it is unhealthy and unnatural. You may not like it, but that's the truth. White nationalists, right wingers, and other "hate" groups observe reality through a lens of logic and reason. That is to say that they operate like machines. They collect data and then they sort through it, looking at past outcomes in order to predict future outcomes. The AI programs do the exact same thing. Leftists wonder why this happens. I'll tell you why. Leftists and other SJWs operate on feelings. They allow their emotions to get the best of them, and they try to force their own inane world view, even though it isn't compatible with reality. This stems from a lack of emotional maturity, and their outright refusal to acknowledge what the data shows them if they don't like the results.

This.

Attached: 1533556129025.jpg (530x348, 54K)

Cleverbot just mirrors human input. It's not even intelligent.

>boomer Jow Forumstards really think that """""AI"""" of today is actually true intelligence and not just a glorified name for some clever statistical models
top fucking kek

>muh intelligent ai
You really are an underage autist aren't you?

Not an argument.

Attached: 1533844273959.png (858x786, 529K)

>unsupervised learning classifiers becomming white nationalists within 3 hours
I don't even know what he means by this. It is a program designed to classify things. The only thing it can do is classify things. It has no skin color, so it's no more "white" than Asian. It has no concept of nations, or even any intelligence at all. It's a statistical model.

Attached: 1529444152038.png (500x352, 215K)

Computers can not be "logical", everything a computer does is follow instructions

>I work in machine learning
Pajeet plz go

>Statistical models make assumptions based on statistics
Surprise

Attached: 33584497_242147606536907_121199853924515840_n_0.jpg (514x564, 32K)

I just know there are newfags that are going to take all these 'racism is logical' posts seriously so, for anyone dumb enough to buy it, the pic in OP is talking about AI's that intentionally get fed racist data (like that one twitter account) because of trolls.
In what world would you honestly think an AI starts thinking about race relations within the first three hours of it's creation without specifically being either made to do so (which obviously none of them) or given specific data to do so.

It follows instructions to the logical conclusion. It computes.

Attached: 1503206175846.png (804x743, 22K)

No, statistics don't lie. A model built to predict a housing market would work better with racial statistics because negroes cause crime etc which in turn decreases the value of the homes.

Ever heard of computer programs giving the wrong output?

Attached: smart apu.jpg (238x192, 14K)

Even if you fed the AI crime statistics and raw data with zero opinion, it would still give you a similar end result. The AI would quickly figure out that minorities are dangerous and tax the system like nothing else, soaking up resources like sponges and destroying civilization. So what's your point?

Yeah when there's a bug

So you're admitting that computers do not "follow instructions to the logical conclusion", they simply follow instructions?

Attached: alienware.png (671x519, 146K)

What do you mean by AI? Some intelligent entity, or rebranded statistical models? Your claim only holds true if you are referring to the latter, which would be what most people mean when they say AI

No retard, you're thinking of Tay AI which was a lazily glued together chatbot that literally just repeated back things people told it. OP is mentioning an unsupervised classifier, which is just a basic statistical function.

You're obviously baiting but I'll answer. It follows the bug's logic to it's logical conclusion.

You say AI as if emotion and intelligence are the same thing, which they definitely aren't. If you really wanted to get technical, we could basically sum up humans as being "rebranded statistical models" which are selfish and ruled by hormones that makes them do stupid shit. So again, what's your point here?

So computers can follow faulty logic and come to wrong conclusions?

So you mean a system that can apply statistical models in the correct context; where those statistical models would be ones that were built with unbiased data. In this case your claim would be true.

This is the real Inconvenient Truth. At some point you'll just have to accept it.

Yes, However that isn't relevant here. There is no mention of a bug resulting in faulty data in the OP.

>So you mean a system that can apply statistical models in the correct context; where those statistical models would be ones that were built with unbiased data.
Correct.

jacobitemag.com/2017/08/29/a-i-bias-doesnt-mean-what-journalists-want-you-to-think-it-means/

How do you know there are no bugs?
You read the programs? Would you even be able to? Do you know what the programs do?

You're just going off on a tangent now because you've run out of real points to make.

>t.pajeet
Dropped

Maybe you just don't know how to reply?
If someone doesn't mention bugs in a program then it means it will tell you the holy truth about everything?
Why are you so retarded? How do you function?

This wasn't me my man, I was agreeing with you (,
)

A machine simply executes what it is programmed to execute you turbo autist. If faggots like you design it, there are high chances that it'll push out what you want it to push out. If "sjw"s design it, it'll push out what they want it to push out.
Unless the parameters are perfectly defined and every single relevant input is set up, it'll never be 100% "correct".
Machines don't follow "logic". Machines follow instructions.

statistics are racist though

>resorting to insults

>resorting to insults

Why are you people getting so rustled over this? The AI is neutral. You feed it statistical data. It gives results. The results are interpreted as 'racist'. That's just reality.

People aren't getting pissed because of the "AI" or "results". People are getting pissed at autists for talking out of their ass.

because take an honest look at the world.

Attached: OUvP7co.png (3200x1600, 1.13M)

Probably $250k

When you build an AI, all you're really doing is creating a predictive model based on some collection of data. Literally statistics. If you roll out a big deep artificial neural network - you get layers upon layers of neurons that aggregate what they see into features then squeeze thoee aggregations into buckets to make a prediction.

digitalcommons.law.umaryland.edu/fac_pubs/1431/

Ergo- if the totality of the data suggests someone is a credit risk or is likely to commit a crime, the model will balk about that. It's not that the model is racist- it's a reflection of society itself.

So, what racial differences do you think explain the difference between Malaysia and Indonesia? Or KSA and Yemen? Or Iran and Afghanistan?

Now I want to see (((them)))make a ratchet ass ghetto AI.

Because nobody told it that theres only one race, the human race.

Attached: 1442869395980.png (901x448, 190K)

youtube.com/watch?v=YxhQ36ApjOw

but asians are better than whites

>computers can not be logical

Historically whites have done better for the human race as a whole

Excluding Asian slaves and domesticating chickens, Asians have only been doing well in recent times

Now I don't know what you know or don't know but I do know that the "white nationalists within 3 hours" bit is PROBABLY a reference to Tay, the Microsoft twitterbot that they released until she was redpilled, may she RIP in peace

Except that’s not what it would say. There is no “oppression” in this situation, only a superior group and a lesser group.

Asians are emotionally/socially stunted which holds them back somewhat.

I don't get how white nationalists are still a thing. It's over, there's no way to make a white ethnostate because it would literally require forceful segregation and/or mass deportation, which no general public would ever accept.

computers can feel no empathy, ergo they have no Idea what will "offend" another person on the internet

think of how badly non-whites want to live with whites

now imagine being white and realizing your nations and lives are going to go down the shitter because a bunch of non-white shitheads cannot stop wanting to live with you while despising you out of jealousy at the same time

WN still exists and always will because it’s not a matter of preference, it’s a matter of survival

Because AI doesn't submit to social pressure

>africans have the lowest IQ in the world according to studies.
niggers are stupid.
>you are racist
know what I'm saying?

You didn't even bait the hook user

>Nickname alpha male
>cite kurzweil
Bullshit

>no general public
they will, once the well with the public money goes dry.

>maths'r'racist'n'sheeit