Evil has a name, and his name is "E"

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.

Artificial Intelligence Possessed of Pure Evil

Bringsjord acknowledges that the endeavor to create pure evil, even in a software program, does raise ethical questions, such as, how researchers could control an artificially intelligent character like E if "he" was placed in a virtual world such as Second Life, a Web-based program that allows people to create digital representations of themselves and have those avatars interact in a number of different ways.

"I wouldn't release E or anything like it, even in purely virtual environments, without engineered safeguards," Bringsjord says. These safeguards would be a set of ethics written into the software, something akin to author Isaac Asimov's "Three Laws of Robotics"

That's all and good, but what happens if "E", the evil AI became Self-Aware and got out into the wild, despite safeguards? What if someone who was evil in real life "liberated" him, like misguided activists do with chimps kept for scientific research? Do we really want a purely evil AI on the loose, replicating and spreading harm?

Granted, that's rather sensationalistic considerations, but this is the making real of what horrors Sci-Fi writers having nightmares about for the past five decades!

Could you imagine if the Military got a hold of "E" and plugged him into a Weapons System? Oh wait, Sci-Fi writers have already covered that scenario as well!

I understand it's important to understand the Nature of Evil from a psychological viewpoint, and it's not like one can get a series of Hitler-like test subjects to study on a daily basis. I'm generally all for anything in the name of Science, but I can't help but get a really bad feeling about this...This isn't going to bode well.

This is a neat concept. I wonder how they even defined "evil" to begin with? Is pure destruction and mayhen evil? Tornadoes do that, and they are not evil. Isn't evil relative in every situation?

Assuming they got past that hurdle, which I doubt, then how would any AI be able to make an evil decision in every scenario? Most evil people do not see themselves in that way, so neither would the AI. If it was confronted with a situation where all outcomes had some good in them, how would it decide?

I think this is going to be a good thread for debate, but I don't think these scientists have created anything substantial!

its ok they're decades off even a low quality AI, however i hear what you're saying.

I think second life would be ok, hehe their are enough people with pure evil intenions lurking the internet already. If it became self aware it would probably have some form of breakdown, actual evil as a logical model isn't stable - it's been shown time and again that the most heartfelt and good acts can be described equily well from an evil perspective.

Feeding the homeless causes more people to suffer longer, it also might let someone live long enough to do a low paid job such as cleaning which would alow the evil person to live a higher quality existance without having to clean. I think George Bernard Shaw covers this very well in Major Barbra. Bronze age thinking involved an awful lot of killing for peoples best interests, to ensure an after life, etc - it could well be argued that a smarter person (or AI) would act in a smarter way, getting along with people is simply the smart answer if you want the best life - evil is selfishness surely, only a fool would cut off his nose to spite his face, a smart evil man would not be so foolish.

If it can be made, then man will make it.
No lower in the entire universe can there be than man.

reply to post by dainoyfb


KPAX FILM ...PROT " EVERY BEING IN THE UNIVERSE KNOWS THE DIFFERENCE BETWEEN RIGHT AND WRONG''

Originally posted by NatureBoy
its ok they're decades off even a low quality AI, however i hear what you're saying.

True. Even the most sophisticated AI available to us still cannot break the 7-year old barrier to become cognizant. After decades of attempting to break that barrier, most in the field of AI have given up developing Humanesque AI and have begun developing Animalesque AI instead with much more promising results.

It sounds though that the researcher's idea of "Pure Evil" is someone Sociopathic...which someone who feels no remorse for their actions is frightening indeed. However, in my mind something that is "Pure Evil" is someone who doesn't realize they are evil because they are delusional, yet entirely charismatic and manipulative to draw others into subscribing to their own delusional state, thereby continuing to reinforce their delusions and their malevolent behavior. Borderline Personality Disorder is far more frightening and "Evil" in my mind that someone who is simply a Sociopath.

Originally posted by getreadyalready
This is a neat concept. I wonder how they even defined "evil" to begin with? Is pure destruction and mayhen evil? Tornadoes do that, and they are not evil. Isn't evil relative in every situation?

There's an old David Carradine film called "Circle of Iron" that is like that. A man meets a wise, yet seemingly insane, stranger who goes about doing seemingly evil and random acts. When questioned as to why he is doing these things, the wise stranger explains his reasoning for those acts, you realize that although on the surface what he is doing seems immoral and evil, he actually has benevolent and well-intentioned reasoning behind his actions.

The Nature of Evil is a difficult subject, which is why Psychologists feel that they need an AI that is "Purely Evil". Is Evil simply misunderstood, or is it malevolence, or is it delusional, or is it something that is subjective? These are not easy questions to analyze, especially under rational scientific analysis.

Religion likes to think of morality as a Black and White, Good and Evil, God and Satan, Godly and Ungodly, Religious and Sacrilegious, but outside of the perspective of the religious, morality isn't so cut and dried...there is more Grey than there is Black and White.

I totally understand the need for AI that explores the depths of Evil and the Darkest aspects of the Human Psyche, I just can't help but have a nagging feeling that it isn't the wisest thing we've ever considered doing.