Study: Humans Show Racial Bias towards Robots of Different Colors

Megalith

24-bit/48kHz
Staff member
Joined
Aug 20, 2006
Messages
13,000
Researchers from New Zealand’s University of Canterbury claim that people display different racial biases towards black robots and white robots. Apparently, people can’t help but “perceive robots with anthropomorphic features to have race,” which results in the prejudices that humans experience extending to robots.

Our study revealed that participants were quicker to shoot an armed black agent than an armed white agent, and simultaneously faster to refrain from shooting an unarmed white agent than an unarmed black agent regardless of whether it was a human or robot. These findings illustrate the shooter bias towards both human and robot agents. This bias is both a clear indication of racism towards black people, as well as the automaticity of its extension to robots racialized as black.
 
Last edited:
Before this thread gets shut down, I'll just say, it makes sense. Regardless of the race issue, people are going to project their own world views on anything they see. For those of you that saw the Silicon Valley episode where they don't want to harm the beautiful young female AI robot, do you think they would have had the same motivations if it was ugly and/or of a different race than their own, or even their same gender?
 
Unfortunately the paper is hidden behind a paywall. I'd like to see the methodology and how the black and white robots compare visually.

That said, I'd be hesitant to draw the same conclusion and even more hesitant to assign a high probability. Whether or not there is a racial preference at play, which is admittedly possible, it's almost certainly not the only factor. A brightly coloured robot is, all else being equal, going to appear less frightening than a dark one. Darkness implies stealth and threat. There's a reason SWAT uniforms aren't flesh coloured. Neither would one expect flesh coloured tactical or military bots while a significantly dark bot could jump out as dangerous.

It strikes me that the data gathered would be more meaningful if they authors had attempted to establish various controls. Eg. responses to various non-flesh coloured robots of similar brightness to the flesh coloured one.
 
Before this thread gets shut down, I'll just say, it makes sense. Regardless of the race issue, people are going to project their own world views on anything they see. For those of you that saw the Silicon Valley episode where they don't want to harm the beautiful young female AI robot, do you think they would have had the same motivations if it was ugly and/or of a different race than their own, or even their same gender?

I agree, but I wouldn't say they're necessarily views, as I feel a lot of stuff like this comes down into a natural bias. And there are reason for these biases on a grander scale survival/evolutionary stage. In the short term, we have to fight our very innate nature to create a more just and fair world.
 
how the black and white robots compare visually.
1118roboteconomy01.jpg


vs

NECA-Toys-Valerian-And-The-City-Of-A-Thousand-Planets-Figures-4.jpg
 
Unfortunately the paper is hidden behind a paywall. I'd like to see the methodology and how the black and white robots compare visually.

That said, I'd be hesitant to draw the same conclusion and even more hesitant to assign a high probability. Whether or not there is a racial preference at play, which is admittedly possible, it's almost certainly not the only factor. A brightly coloured robot is, all else being equal, going to appear less frightening than a dark one. Darkness implies stealth and threat. There's a reason SWAT uniforms aren't flesh coloured. Neither would one expect flesh coloured tactical or military bots while a significantly dark bot could jump out as dangerous.

It strikes me that the data gathered would be more meaningful if they authors had attempted to establish various controls. Eg. responses to various non-flesh coloured robots of similar brightness to the flesh coloured one.
This was my immediate thought as well. Studies have shown that different color wall paper causes different emotional reactions. The wrong color in prisons, for example, has been shown to increase violent behavior. Isn't it possible that black, as a color, just has an ingrained human response?

Here's some studies on the affects color has and how colors are viewed:

https://www.verywellmind.com/color-psychology-2795824
 
Last edited:
I agree, but I wouldn't say they're necessarily views, as I feel a lot of stuff like this comes down into a natural bias. And there are reason for these biases on a grander scale survival/evolutionary stage. In the short term, we have to fight our very innate nature to create a more just and fair world.

Natural bias, and environmental factors are not the same thing.
 
Unfortunately the paper is hidden behind a paywall. I'd like to see the methodology and how the black and white robots compare visually.

That said, I'd be hesitant to draw the same conclusion and even more hesitant to assign a high probability. Whether or not there is a racial preference at play, which is admittedly possible, it's almost certainly not the only factor. A brightly coloured robot is, all else being equal, going to appear less frightening than a dark one. Darkness implies stealth and threat. There's a reason SWAT uniforms aren't flesh coloured. Neither would one expect flesh coloured tactical or military bots while a significantly dark bot could jump out as dangerous.

It strikes me that the data gathered would be more meaningful if they authors had attempted to establish various controls. Eg. responses to various non-flesh coloured robots of similar brightness to the flesh coloured one.

It was a joke. Really. They used fast paced computer stimuli of images where a choice was made, to try and replicate the fast paced nature that police officers have to react to. If you see the images its a joke, they are not standard. For one, the "white" subjects, as they used humans first before showing the "robots", the first white human shown is facing SIDEWAYS, they have a phone in their hand, but in a position as if they are answering it and putting the phone to their ear. The black human shown? He is FACING the camera, and while he has a phone in his hand as well, his arm is down and his elbow and forearm is straight out in the same manner someone would POINT A GUN. In fast paced situations body language and the simple fact that the white person if not even facing you are going to be the main risk factors your brain is thinking about first.

The second set of human images also shown before the robots are the armed images. Again, the black human has the gun in hand pointing outward like someone would do who is going to shoot. The white human has the gun in his hand, but his arm is straight down, not pointing at anything and you cant even make out the profile of the gun very well.

The very fact they showed white and black humans before showing the white and black robots show that they wanted people to assign a race to the robots. The fact that others at the convention where the paper was presented to peers, a majority of them claimed the paper was "sensationalized".

The "significantly faster to shoot black" is also stretching, as the time in ms to shoot an armed white subject was 605.88ms, while time to shoot an armed black was 601.22ms. Which is pretty small when you think about the way the images were portrayed.
 
It was a joke. Really. They used fast paced computer stimuli of images where a choice was made, to try and replicate the fast paced nature that police officers have to react to. If you see the images its a joke, they are not standard. For one, the "white" subjects, as they used humans first before showing the "robots", the first white human shown is facing SIDEWAYS, they have a phone in their hand, but in a position as if they are answering it and putting the phone to their ear. The black human shown? He is FACING the camera, and while he has a phone in his hand as well, his arm is down and his elbow and forearm is straight out in the same manner someone would POINT A GUN. In fast paced situations body language and the simple fact that the white person if not even facing you are going to be the main risk factors your brain is thinking about first.

The second set of human images also shown before the robots are the armed images. Again, the black human has the gun in hand pointing outward like someone would do who is going to shoot. The white human has the gun in his hand, but his arm is straight down, not pointing at anything and you cant even make out the profile of the gun very well.

The very fact they showed white and black humans before showing the white and black robots show that they wanted people to assign a race to the robots. The fact that others at the convention where the paper was presented to peers, a majority of them claimed the paper was "sensationalized".

The "significantly faster to shoot black" is also stretching, as the time in ms to shoot an armed white subject was 605.88ms, while time to shoot an armed black was 601.22ms. Which is pretty small when you think about the way the images were portrayed.

Of course they did, I was skeptical upon reading the absolute pronounment that “there is rascism you are a rascist X X X” but thanks for confirming it. I truly want to believe there are more good than bad folks in the world and so far for 37 years I’ve seen - I’m probably just lucky, but have lived everywhere from Oakland to Vancouver to Cincinatti - mostly good.
 
I though all robots were supposed to come with a digital lights controller ???
 
some call it learned response to real life scenarios. But referencing real world stats on crime would brand that person as racist as well.

now, please lets do a similar study other countries in other than those where the whites are NOT dominant.
 
Always nice to see that experiments mirror real life results.

I've been warning a couple of friends that they should never ever get drunk. That they should think about the possible worst scenario that can happen if they do. Of course, in the current political climate, my inclination is that they have nothing to worry about.
 
I hate robots so yeah I'm racist as fuck against stupid retarded non human synthetic dumbshits replacing jobs and crap as we dissolve into meaningless no longer needed rejects of the technocratic overlord despots.
 
Ironic,since studies have also shown (from actual, real, science colleges) that our police forces are LESS likely to shoot people of color at rates statistically fantastical from the shooting of white people.
 
Sure there is a racil bias! The red ones are designed to rip off your limbs, while the green ones lull you into a deep coma, and the purple ones grind you up for the minerals you possess, while the yellow ones suck out your eyeballs.

In the end, no matter the color, they are going to KILL US ALL!
 
They totally disregard other possible things that may trigger defensive responses, such as it is harder to tell details on robots that have dark colors because of a lack of contrast, so it is harder to see if one is armed or not. There are also others, but due to it being just before 6AM my time, I'm not prepared to argue. Still, half asleep or not, I can tell BS when I smell it.
 
Researchers from New Zealand’s University of Canterbury claim that people display different racial biases towards black robots and white robots. Apparently, people can’t help but “perceive robots with anthropomorphic features to have race,” which results in the prejudices that humans experience extending to robots.

Our study revealed that participants were quicker to shoot an armed black agent than an armed white agent, and simultaneously faster to refrain from shooting an unarmed white agent than an unarmed black agent regardless of whether it was a human or robot. These findings illustrate the shooter bias towards both human and robot agents. This bias is both a clear indication of racism towards black people, as well as the automaticity of its extension to robots racialized as black.
He's a freaking white robot!!! Obviously it's wacist!
 
I don't care what color our future robot masters are as long as they have tits!

NSFW! YOU BEEN WARNED....






 
As a brown skinned racist I find the data draws no conclusion since dark skinned Asians and Hispanics were left out to continue pushing the fake racism agenda.
So this is what a higher education is comprised of.
Glad schooling never got in the way of my education.
 
Wait a second, are we talking about sexy robots or regular robots? Because.... reasons.
 
So what about comparing black colored robots with red ones, or adding neon green to the test.
What was the color of the background? Did the white or black robots blend in better?
If they are going to make this a study of racism, how did the race of the shooter effect the results. Were black shooters more or less likely to shoot at black robots? I'm curious if considering the color black to represent a greater threat specific to a particular race of people, or if it is more universal. Is it an us vs them bias, or a more general bias.
 
Back
Top