Slashdot Mirror
Team Aims To Create Pure Evil AI
puroresu writes "Scientific American reports on the efforts of Selmer Bringsjord and his team at the Rensselaer Polytechnic Institute, who have been attempting to develop an AI possessed of an interesting character trait: pure evil. From the article, 'He and his research team began developing their computer representation of evil by posing a series of questions beginning with the basics: name, age, sex, etc., and progressing to inquiries about this fictional person's beliefs and motivations. This exercise resulted in "E," a computer character first created in 2005 to meet the criteria of Bringsjord's working definition of evil. Whereas the original E was simply a program designed to respond to questions in a manner consistent with Bringsjord's definition, the researchers have since given E a physical identity: It's a relatively young, white man with short black hair and dark stubble on his face.'"
I suspect that they(ironically detracting from their goal) went down the path of maximising for "threatening" or "untrustworthy", rather than evil(which is much harder to depict, without falling into specific cliche-riddled stuff).
A fair few studies suggest that a face that looks about like that one, with more or less unpleasantly masculine features, rates low on perceived trustworthiness and high on perceived threat. Of course, the evil that you don't recognize is way more dangerous than the obvious one, so choosing that is kind of silly; but I'm not too surprised that they did.
Isn't it wonderful to be in the world today, where everyone can be racist against whites without fear of reprisal?
So to summarize...nobody is the villain in their own story.
"Never let your sense of morals prevent you from doing what is right" - Salvor Hardin