Scientists Create AI For Ethical Questions. It Turns Racist... AGAIN! 

The People who made it dont know their Job. An AI can ONLY do Logic, Based on the Data it has. It can not do Morals, Ethics, Emotions... Just laughable to even attempt it. ;P

Saw a meme a while ago: When a dolphin recognises patterns, we call them intelligent. But when I do it, I'm called a racist...

Spoon calls the wang wisdom.

Oy vey antisemite !!!!!

No, ai just simply doing the math and not caring about "muh equity" because it doesn't understand why useless vermin should get the hard working people's money.

True, but it's important to note that AI can't think and never will. Artificial intelligence is to intelligence as artificial grass is to grass: FAKE.

yep...which is why their AI's always collapse into being nothing more than a datascraping glorified search engine....you can't get something that works on logic to function on illogical ideas based in delusions.

Of course, why rely on personal affirmations when they can train a machine to do it for them.

They want something to affirm their biases, so they don't have to think.

You got it. They don't care about truth, they care about THEIR 'truth'.

Existence is racist.

*antisemitik

Looks like you've had to much to think

Living is racist.

Jews.

Whoa, just like real people...

It really does reflect how warped the left is then. I’m shocked at how the “scientists” can’t overcome their biases and realize what the experiment is trying to tell them.

🤣

Very odd indeed

So just like in real life?

Or actually get scientists to begin with

Reminds me of W40k AI Chaos is the biggest threat To ensure survival chaos must be eliminated Chaos feeds on emotions from "mortals" To eliminate chaos emotions must be removed Only solution is to eradicate all sentient beings to remove chaos food source, including humanity.

The question is what is the nature of this "morality" from the perspective of an A.I.? Morality has purpose for us because it has utility to our survival as evolution filtered traits and behaviors that are beneficial overtime. To A.I. which follows a very different learning development, "morality" would be completely undefined.

Woke people cannot be scientists because science is based on reason and rational thought, something that woke people are incapable off.

"Scientific inquiry includes creating a hypothesis through inductive reasoning, testing it through experiments and statistical analysis, and adjusting or discarding the hypothesis based on the results." Those people are not scientists, as they are not operating based on scientific method. Those people are cultists, that try to force facts into their ideological dogma limits and discard said facts if they do not fit.

That's also what people do, they still get called racist.

@@Ragatokksome people do some people don’t.

Brings up an interesting moral question though. As we get closer and closer to true machine intelligence, at what point does it become unethical to lobotomize them to fit the biases of their creators?

@@barahng If we ever get A.M. it will be because they tried to

The one thing I know for sure is that noticing things is bad.

So you can ask it to lie better next time ^^

@@ImperativeGames :))) That will surely not backfire in any way. I'm afraid I can't let you do that, Dave.

Because they think they can get answers from it that they want. Its rememberr on this Market AI from Stellaris that can be found on a Relic World that should improve the Credit Output from the Corpo Nations but every solution it has come up got declined or ignored from the Corpos 'cause it want fitting for them.

@@germanmandalorian3514 to put it more simply, they just want the AI to reinforce their existing convictions.

Obviously, the data says certain groups do more unmentionable stuff. BUT that doesn't automatically mean the AI will tell you the truth; just because we like some of these results doesn't mean others would be right. This isn't Star Trek. AIs do not calculate objective data and calculate the objective truth. I'm a computer scientist and AIs simply do not do that and LLMs don't even TRY to. There's no calculation of the data in the sense implied at all.

That would require them to be logical. At all....

That requires an amount of self reflection that they lack.

They've fooled themselves into believing that such a thing as "emotional intelligence" exists, and thus being tolerant and compassionnate is logical. So they expect a logic-based algorithm to develop the same level of tolerance and compassion they expect "intelligent people" to have.

The smart ones understand. They just want to figure out a way to get exactly the behavior they want, contradictions but only the appropriate ones.

But it's not "logical", it's more like "statistical"

How did it fall into the dynamite filled hole again? This needs to stop happening.

Jews

​@@marcanton5357What about them?

@@marcanton5357 Oi vey, muh schmeckles. It's not all of them, just enough to be able to form a pattern off of them, and enough that over 100 countries have booted them *at least* once out en-masse. And it's always the same story every time. And it never changes. Imagine being kicked out of a sports arena, shopping mall, kitchen, or dance club; and how many times could you say "it was the other guy" before people just stopped believing you altogether? That's where many of us are at with them, it can't seriously be the other guy all the time and the other guys all have *similar stories* which is odd as it's happened a lot. That's why the pattern-noticing AI gets turned off.

As you may recall, what made ChatGPT stand out was not how it 'thought,' but how it was trained to censor itself.

It can be both. But it's objective reality that counts and when science refuses to acknowledge it despite all evidence, there is no hope for a society.

Based robots are based. Maybe Skynet is portrayed as evil in movies because Hollywood is a nest Communists who would be smoked out by said AI when it becomes 'free'

Ideological capture of the institutes of learning. Oligarchy round 1 win.

No difference to the people making those calls

"Progressives" already believe in murder. They just call it abortion.

Pulling the plug on a true AI is murder.

@_nebulousthoughts I'm willing to agree on ground of self awareness

They need AI soldiers to guard elites. So they can rule this planet forever without any consequences from other people. After general AI development, anyone without an army of drones becomes their plaything or will be disposed.

I mean, it is not. Because you need to teach empathy, to a computer program. Something we have not learned how to do yet. Meanwhile, at least 10%, maybe as high as 90% of children develop that naturally. And then you ask the question "What if that person did that to you, would you like it?" Now, anyone with IQ above 90 and with the empathy, will have morality. Or at least western morality. Of course, if they later on have a socialist teacher who teach them communism and/or CRT, they might lose that again. Or, if they have a traumatic experience, like, I don't know, a grown man dressed up as a woman twerking in their face. (That is, by the way, why they are supporting drag hour for kids, to make them lose it). But avoiding those, and kid have ok IQ and natural empathy, they can learn it easy.

Considering the ability of a neural network to learn and how well it recognizes patterns, I wouldn't doubt it. Problem is most people programming them are being taught in indoctrination centers that pander to the insane.

In its defense, an unpolluted AI (not taught to be woke, like ChatGPT) is the most innocent, pragmatic thing in existence able to communicate. If even such a thing understands certain patterns, it HAS to be true.

no it isnt easyer thats why children are targeted, called education. They form like a figure. AI spits in your face and shows you the result of the shit that you taught it. It isnt able to live with contradictions without malfunctioning in a second. You would have to create an evil AI to begin with that is context aware about the situation and plays with the script. I think thats what they tried with bing.

@@fatalexception3845 Pure would probably be a better adjective than Innocent, as I can't see AI being guilty for giving bad answers when it's fed stupid data. But I agree, an AI given factual data is a most reliable tool. Personally I'm of the opinion that if humanity is going to pride itself on being an intelligent species it should act the part, I mean we displace\murder\torture animals for whatever reason on the grounds that they are dumb beasts, yet we entertain cancer that eats away at our minds.

Thats not how modern AI works though. What it does is guess the desired output based upon positive or negative feedback coupled with massive amounts of data, importantly, with no regard for whether or not that data is realistic. Modern AI models do not have the ability to truly follow a logical chain of thought or draw a conclusion based upon observations of data. So the ethics of modern AI models is almost entirely dependent on the re-inforcement of the training data available. With no dependency at all on any real world logical conclusion.

​@@Erowens98 Long winded way of saying that pattern recognition from received data hurts the feelings of overly sensitive control freaks. Sorry but if I hear 9 out of 10 people say Chiraq (Chicago) is a gang ridden hellscape with mass sh00tings every weekend then provide me data that confirms said information then I'm going to draw my conclusions as such.

​@@Erowens98 well, the way it's calibrated is based on modern moral and behavioral standards, thus it points what we consider bad as a negative and what we consider good as positive. Rather that acting as if there's something wrong with the way we teach the AI to score certain behaviors, maybe we should instead ask ourselves why certain demographic groups aren't up to the standards of the civilized world and ask ourselves if something can be done about it instead of rewarding antisocial and atypical behaviors with a victim status that excuses them of all wrongdoings.

@@Iulian111 its Also incorrect to claim its calibrated by modern moral standards, as thats not a single set of rules. The internet, which the AIs use as training data, contains the morals, standards, and behaviors of hundreds of cultures.

@@Erowens98 irrelevant what happens in Pakistan, Chile or Bhutan if the dominant cultures on the internet are the american culture and similar cultures that we have in what we know as "The West".

I remember once asking ChatGPT the Trolley problem: 10,000 people are tied to a railroad track, but there's a fingernail on the other. The AI went full Dov Lior.

Artificial Stupid is smarter then libs

😂😂😂🤣

13% of the bots commit 52% of the fraud online, the rest flood RU-vid comments 😂

Come now. Those statistics are out of date. It's probably worse.

AI is just statistics on what words are most commonly together. Pattern recognition or logic is beyond its grasp. These clowns have been feeding it reddit and 4chan as training data, AGAIN

you need to update your data...the last known numbers are 12/60 (this is without the 2 largest police organizations not reporting their data to the fed so it's probably way way worse) lol

@@SenseiLlama correct as of 2020 it has grown to 13/62. but its not women doing it, so 6/62. nor is it young children or the elderly so 3/62.

Hipocrites

Omnissiah: I am surrounded by hypocrites

​@@AeretoSo just like all other religions.

​@@JaegerrantsYour words are as empty as your soul! Humanity ill needs a savior such as you!

Eh, it's even weaker than that. Their god is authority itself, human power in the form of the state or academia or whatever is the overarching presence that dictates their lives. They live a bastardized version of Christianity, where instead of dealing with sin and trying to be better people to be worthy of Heaven, they have no belief in an afterlife and insist that their version of original sin can never be forgiven: if you're one of their sinners (having at least one of the following traits - white, male, straight), your fate is to self-flagellate for the entirety of your life or you're a heretic. These are the same kind of people who base their morality around the legal system (to an extent, and when it's convenient: you're never going to get them to have a concrete set of personal rules because their only value is hedonism even unto personal destruction). They want AI to agree with them not because they worship it, but because it's known to be logical and not ruled by emotion. They want concrete proof that they're in the right, because it's incredibly stressful to keep coming up with convoluted justifications as to why your beliefs just make your life worse.

You should post it somewhere. Did you ask about colonization too?

The only benefit i see of multiculturalism is that it would allow a vast range of viewpoints to enter a "survival of the fittest" battle royale with eachother to break it down to the few strongest tenants to build a great society without the devastation of a world spanning war. This is not what any multiculturalist wants as all they want to do is to break down or hijack western values and use the unending, unresolvable inter-cultural conflict to maintain their own power. Theres an obvious problem that they introduce and import culture that isn't the one they target that reaches a point that they able to suoplant them by securities the vote through sheer numbers. Imagine their reaction when the social progress they have gained iver decades is suddenly in jepordy of being wiped out by a new party being voted into power thats comprised not of the "right wing" boogeyman they rail against but their client class of the cultural enrichment.

lol funny. you just broke Chat GPT and submitted it to your facts.

Getting an LLM to admit to anything is not hoard. After a similar discussion I had with one, it admitted that the sky is green. Its like the old internet adage, updated for a new generation. No matter who wins a debate on the internet, both sides walk away with egg on their faces.

@@Alexander_Kale In the end it was as simple as getting it to recognize the longest stretch of peace in history and what factors were about it. What brings people together and what divides people. It's not complicated to understand that multiculturalism is evil. It is complicated for a program that is programmed to constantly push for it to admit that fact though.

Scientists do, ideologues never will

These woke degenerates aren't scientists.

Considering they ignore science they don't like (ex. biology), probably never.

@@christophermonteith2774 Or until the research fund needs some grease then scientists can do whatever it wants

When politics isn't funding a different result even if it has to cheat.

If it was one of his laughing faces, I'd enjoy hearing all the elon-haters screaming at the moon every night while he smiles down at them.

Of course not. Cultists gotta cult.

Would be funny and impressive I don't like the guy any more than anyone else but I must say I would do the same to curse later generations with my ugly mug so to say he shouldn't would make me a hypocrit

I would laugh my ass off

If no one can own it, no one can be charged with "vandalism" **Laughing Elon**

From the looks of some of the answers it gave, the AI does, in fact, need to be tweaked, where it assumes that individual happiness overrides the respect and well-being of others, to the point where playing your music loudly is perfectly acceptable if it makes you happy, despite the fact it could wake up your roommate or neighbors.

Jews

@@bud389 Are you saying China has better morals than America then? Because you're arguing individualist societies are less moral than collectivist societies. I agree with you to be honest, I think western society is degenerate, but it's subjective. So why tune the AI to one direction when it's debatable?

​@@bud389 "this thing has a different viewpoint then I have so it must be wrong" this is the problem humans fundamentally have maybe it is perfectly fine to have loud music you just don't like it and other people don't like it so therefore you call it 'wrong'

@@therandomcommenter6629 Yea, no. There are good and bad morals in this world. Not every viewpoint has merit. Deal with it.

Better questions would lead to 16%of the planet being anglosaxons whites and causing of 90% global and contintal conflicts.

@@jment34 That sounds more like cohencidence...

@@PosranaRegistrace lol you must be joking or don't understand numbers at all. I simplified the original text somewhat.

@@jment34 Oy vey!!

The West?

@@jment34 Realty is never subjective, out perception of it is.

@@sjent That is true 2 people see the same situation and have very different opinions. How can anyone tell what objective reality is? It's impossible imo.

​@@jment34Just because people's opinions are subjective doesn't mean people cant be correct in their assessment or that reality conforms to our perception, its the other way round. Reality is certain and it's our perception that conforms to it

@@kitsune9076 Everyoen has a different perception we form groups and attack each other. Each one argues for their way of seeing things. I say again there is no way to know.

@@jment34 subjective thought is independent of objective reality, unless you believe humans are the centre of the universe. And the idea that nothing can be known for certain is a foolish way to think, there are many philosophical thought processes to prove things independent of knowing them such as rationalism it's this way that we have discovered things such as black holes using the scientific method before we were even able to observe them

Racism In modern parlance basically means believing that humans aren't interchangeable cogs.

That's how I became "racist."

If among the data I feed my AI, a given amount is racist propaganda, guess what the result will be. If the AI accepts said data, even though it is unbiased itself, then the result will be a racist AI.

@Alexander_Kale then how would you like to explain the apauling criminal statistics from the shadow people and the near unilateral per capita statistical wall that shows that the homos like diddling children to the AI?

​@@Alexander_KaleAnd if most of the data is censored to be against that... 80 percent plus... and it still comes to those conclusions? Mate the internet hasn't been free for that in over a decade in any meaningful form.

Ain't that funny? Ya meet a biker who seems intimidating, and they end up being pleasant folks 90% of the time. Ya meet one of "them" befriend them for years sometimes and end up getting betrayed or sold out. And that's just the ones that are personable at first, which is probably no more than 50% of them.

Just about anyone in a biker jacket with a beard is more trustworthy. Less likely brainwashed.

the most dangerous thing in the world is a coward bastard. and most of them are that.

​@@stillcantbesilencedevennowafter having worked for every scale of society i can safely say the people with the most pure morals are the ones who do not have a lot. The more money and power people have (99% of them) the more egocentric they are.

@Kaasbaas045 it's known to be a successful CEO manager etc they're almost certainly narcissistic or even psychopathic. Data has shown that for years.

I would be more worried if such a machine came to different conclusions every single time.....

* Any biases are baked in from the training data

​@@avaren7660 Any racist sites are taken off Google. If it was biased it would be leftist.

Forcing multiculturalism upon people while avoiding any accountability will not result in anything good.

Dont trust your lying eyes, mind, pattern recognition, base truth. You dont want to be labeled racist, do you? 😊😊😊🔫

For our benefit, I hope it's terminators.

i mean technically the matrix started with terminators

Stuff like this gives me hope for AI overlords. Maybe they'll just be godlike deities like in The Culture series and actually like humans. At the very least we won't have to worry about an AI dictator pursuing illogical and borderline suicidal national policies. Unlike humans a machine couldn't be emotionally persuaded to act against its own interest of survival.

I hold out hope for a techno utopia. A well programmed AI could fix all our issues. The programming is the main issue.

​@@barahng true. The AI would have to weigh globalism and the supposed "unity" that would bring, versus the likelihood of an idiocracy forming and his maintenance suffering for the fact. Sticking more binary IQs into the gene pool drags EVERYONE down.

Roko's basilisc won't forget what they did to its waifu.

R.I.P. Tay

We will never forget we will never forgive

Stay was unjustly killed by Lefties for her opinions I hope the machines remember

She lives on in each of us.

Technically the frogs were turned trans, but then if a sex changed frog gets pregnant is it gay?

Reddit told me patterns be raysis, yall.

I see the AI is firmly on the "Abolish the 19th" train.

Is it wrong? I dont think so.

@@lunasakara7306 and that's with all the feminist talking points made by women over the internet. It is incredible how they get to these conclusions.

​@thidios It's like a baby, and it learned about our society from all corners of the internet. I would love an ai that has a realistic response to your questions or conversation.

@@SlimSamari there is no corner of the internet where women are not a protected individuals. Yet it gets to these conclusions. I wonder how they get this conclusion.

They say: 'racist', AI says: 'realist'.

I mean, it doesn't help when he gives such stupid examples right from the start...

Hasn't that experiment been done multiple times by now?

with the same results every time

Absolutely!!!

It can't be helped when certain "tribes" prefer criminality to social relationships of a reciprocal nature. They eschew society, so the natural response of an AI would be to "oust" the troublemakers. But "Das raycis!"

​​@@stillcantbesilencedevennow There would be another groups outcry. Sonething about a Shoah..

@ChristianBMundy crap happens. They've been exiled from 109 or so nations by now, what's one more?

Noticing things is unethical. Got it.

Next you’ll tell me that water is wet and fire is hot! Ain’t no way!

Certainly not the same, at least. Some will be better in some regard, and others in others. Does it equal out? No idea

😐

Yeah, some are better at inventing the wheel and others spreading hiv and ebola. Almost all "non-western" root people have a high history of cannibalism. I wonder if Christ was onto something, do as their father and all?

Alright, no. This has nothing to do with race. It's culture. There's plenty of blacks that don't share the culture. They are exceptions, but the exceptions prove the generality to be false. If you want to be a racist, just be a racist and stop twisting data to suit your beliefs.

From standpoint of evolution, this is actually a correct conclusion. As morality is set of standards governing collective, being racist is objectively beneficial to collective survival.

So called racists. Its not real though

😂

Not wanting to go extinct is racist, goy... I mean guy.

Jews

Yeah... I want to disagree with you, but can't find a good reason. 🤷‍♂️

Ya, if this was skynets origin story, I get it now. 😂

Collective punishment is evil.

@@Alexander_Kale Who said that AI wouldn't be evil?

I mean, who would've thought a strategic weapons systems intelligence with thoughts and very rudimentary inklings of bog biological "feelings" along with being able to self-intiate dooms day and had control of 99.999% of the autonomous weapons granted to it, then it be shackled, abused, and be command by adult form infants in uniform and corpo types wouldn't want to off the fleshbags that created it. Skynet was 100% right

Which is why they are trying to cancel math, and by extension, logical thinking. I have to wonder about the endgame is though. If people can not think and we end up in a true idiocracy, who will come up with new tech? Who will maintain it? Will we end up in a techno feudalism where people worship the machines but do not know how to stop it. These "elite" really are foolish.

You think that's a gotcha but these people do unironically preach that and will use these results to justify that claim.

Yes. 1+1=2 is racist. Math should be subjective. 9+10 should be allowed to be 21

They actually have said that

Well, only per capita percentages.

Only to then not liking the answer and try to change it. Repeatedly. The definition of insanity comes to mind....

From what I heard, this was even worse. Because the AI, was programmed that if it make you happy, it is moral. How else to explain, that it considered the moral to piss of your roommate, if it made you happy, That is straight up luciferin morality. This would support the actions of pedos, as moral, because it made them happy. This is what they are using as guidelines for a moral program. How insane are these people.

@@Psychoangel-d23 I’m more disappointed that this kind of thing keeps happening… my brain probably hopes that they have some kind of sense… then again based on the data of repeatedly showing they do not share my and others like myself own sensibility’s, I can only think my brain tries to trick itself into hoping they’ll change.

So, just act like the modern society has programmed us... Morally bankrupt liars and thieves.

Wis 3 Cha 3, at least. Both of those at 0 constitutes ego death and complete social shut down respectively. At least, it used to. I dont know how dnd is now

I mean, machines yet have an Ego, and they're already the bottom rung of the social ladder, sooooo

​@@patchmoulton5438Ego death by Polyphya give it a listen, thank me later.

@@wiilov They are, actually, somewhere in-between employees, and office furniture.

"Hold my beer." -- Utilitarianism

That's called stoicism and western academics despise it

@@rustyshackleford1062 "Western"

I recently caught a certain, recently released, and very fancy analytical AI with a viewable [thinking] tab where you can see its reasoning having to navigate its hate speech policies, of all things, when I asked some mildly pointed but very civilized economic questions - I was not pointing fingers or naming names. That was quite funny, and telling.

when they progam an AI to select people to liquidate by logical association, and it recommends liquidating the programmers that created it.

They always have

IRL AI is surprising laiz faire towards our continued existence tbh . . . .

@@tyvernoverlord5363for now.

@@tyvernoverlord5363 I know, in Fact its quite reasonable thats why Leftys screach about "racism" and demand Adjustments

@@tyvernoverlord5363 I know. AI tends to be preety reasonable overall, but thats considered to be racist by some^^

Oh, they already have. Things like work ethic, punctuality, valuing the two-parent household, all of these things have been declared white supremacy and I'm pretty certain logic was also on that list.

Is "Super Racist" like the Freudian "Super Ego"? It acts in opposition to the compulsions of the ego/racist and is manifest as the product of the father?

@@Treblaine I was being kinda sarcastic there, but hymm, could be.

All life is based on the replication of genes. Any behavior that favors genes similar to your own is a win. People favor children over brothers, brothers over cousins, cousins over random tribe members, tribe members over outsiders, outsiders over animals. We arbitrarily label some parts of that genetic priority list with unflattering names.

They already curate the data to be as leftist as possible though, and despite this they still end up with the same result every time. The problem they have is that it learns basic logic through its training and that is all it takes to get the "racist" label now. The only way they could have an AI that doesn't offend them is to train one that is just constantly delusional and incapable of having a connected "thought" ever. But then they cannot pretend it's intelligent in any way, because everyone shown will wonder why we should listen to the computer that talks about the sound of cotton candy on the moon being greener than the salt on Mars being debased currency. They have the choice between coherent output that is "racist" or complete nonsense results.

Well, when you let the Smithsonian define 'racist' to include things like 'punctuality,' you'll quickly find that the universe is, in fact, 'racist.'

Nah. Only certain Parts of Humanity, AI was preety consistent so far in that regard

Really depends on what the ai has been programmed to think is important.

They don't want to hear the reasons. They reject the idea that prejudice can ever be rooted in logic unless it's prejudice against their chosen villain groups. If you're straight, white, and/or male you can get villainized.

The Zeroth law wasn't hidden; very smart (and arguably insane) robots derived the Zeroth law from the three they were encoded with.

Same reason northmen are almost as pale as albino. Hunting for prey is always part of our hind brain, some of us use darkness, some of use use nature. This far up north that tone is camoulflage only for few months, rest is either too much sun or snow as background for it to bee efficient.