What annoys me most about this test is that the premise is really good! Lets crowdsource human morality metrics for dangerous car situations. Great!
But if something is done badly, this badly, it spoils it for someone who wants to do the same thing in the future but better. People will just think "oh, I already did this test, and it told me I preferred killing criminals and young people over others" and not take part in the better designed test.
What annoys me most about this test is that the premise is really good! Lets crowdsource human morality metrics for dangerous car situations. Great!
But if something is done badly, this badly, it spoils it for someone who wants to do the same thing in the future but better. People will just think "oh, I already did this test, and it told me I preferred killing criminals and young people over others" and not take part in the better designed test.
I'd love to see the data from a higher quality test broken down by demographics of the person taking the test. Especially nationality. How do people in countries where traffic laws barely matter answer this test compared to places where they are strictly enforced?
The nature of these tests is always one of extremes, it puts a magnifying mirror to your hierarchies and values structures. We tend to approach decisions with nuance and often don't have the will or power to make life ending decisions, so we react viscerally when confronted with these structures out of context and hugely magnified. Just because you don't like the results, doesn't make it a bad test.
I put literally no value into gender or social status, but it put me at the maximum for protecting women and high status persons. Clearly the questions aren't properly designed to remove correlations between different factors. Or it works well in general but I'm some crazy outlier. But mathematically, trying to get 9 uncorrelated factors from 13 (IIRC) binary choices isn't a great idea.
One problem with quizz is that it's too short. It sees patterns that happen by random chance. There should be enough questions with enough variations to remove the possibility of that happening.
One problem with quizz is that it's too short. It sees patterns that happen by random chance. There should be enough questions with enough variations to remove the possibility of that happening.
Then no one wants to take it because there's 100 questions.
I protected the people in the car and minimised intervention...
I chose to protect them because they paid for the vehicle. If the customers are the ones that tend to die then the technology stops in its tracks right there and we are denied a future where machines manage all this for us.
Minimising intervention is also important because it allows us to predict what will happen more easily. If I need to predict which lane I should move to in an emergency it's no good have to deal with philosophical puzzles over awareness of physics of the situation.
I protected the people in the car and minimised intervention...
I chose to protect them because they paid for the vehicle. If the customers are the ones that tend to die then the technology stops in its tracks right there and we are denied a future where machines manage all this for us.
That's the most vile and evil thing I've seen in this thread. Money should never be a factor in judging the value of one human life over another. That is literally the root of all evil in our society.
Comments
But if something is done badly, this badly, it spoils it for someone who wants to do the same thing in the future but better. People will just think "oh, I already did this test, and it told me I preferred killing criminals and young people over others" and not take part in the better designed test.
A kid is worth 1.01
A pregnant woman is worth 2.0
Everyone else, regardless of all other factors is worth 1.0.
I chose to protect them because they paid for the vehicle. If the customers are the ones that tend to die then the technology stops in its tracks right there and we are denied a future where machines manage all this for us.
Minimising intervention is also important because it allows us to predict what will happen more easily. If I need to predict which lane I should move to in an emergency it's no good have to deal with philosophical puzzles over awareness of physics of the situation.