@zyx066 thank you! your comments always add another layer of thought to consider! This experiment truly demonstrates the need for a universally accepted standards for the value of life... That is really what morality comes down to... It's clear that life value varies across the globe, and this is the prime issue is driving us to conflict, and also one that humanity will have to solve before delegating that power to AI... and I agree that we might never fully come to accept a single standard... perhaps an average of all the individual standards combined will be the best we can ever do...

Also you made me realize that I forgot to add a very important paragraph that was lost in the spell check... This test might make a person feel embarrassed about their results, because the results don't show what situations you had to judge, and all the nuances you had to consider... sometimes a set is not very balanced and you will end-up saving more thieves than you expected, and then it will appear as though you have preference towards criminals... Its important to not view this test as a personal moral evaluation but rather as that of an AI. Thank you for pointing this out.