The thing that makes me a bit nervous is not the concept of AI so much. But that *we* will program/create the AI. We are, as you point out, rather an imperfect race. We have hates (that make it unlikely that we will be the ultimate survivors on the game of life- unless we change, what else is new?) petty jealousy, irrational behaviour, government;), what have you. When was the last time you created something perfect? AI will have bugs. Any complex system will have bugs. (Humankind has bugs, if you consider sickness, insanity, et. al. as bugs. Although you could consider them evolutionary "kicks in the pants"). If AI becomes pervasive those bugs could well (will probably) threaten life. Even if we didn't make any mistakes (Hah!) what makes you think your goals and ethics for AI (or -more importantly- mine;) would be shared by its creators? Especially considering that those most likely to be pioneers, directly or indirectly (funding), would be Big Government. As in military technology.
On the other hand, I think that "real" AI would most probably be created by "lesser" AIs. Which I would hope would dilute the human element. And we already have human "bugs" endangering life. Anybody up for a bit of human sharpshooting from a high tower?
Definitely an interesting subject, and one well worth thinking about, for unless there is something "magic" about humankind (highly unlikely, in my opinion), AI *is* coming.
Granted, this post could be considered an unlimited resource; but the time, effort, experience, whatever, that led and allowed smack_addict (and later lurker 786;) to create it is not. In other words, the number of posts by smack_addict is not an unlimited resource. Not to mention the number of posts by X moderated up to Y. What about the universe of good posts? I'm sure everybody can extrapolate to other creative endeavours.
Which is not to say I necesarily agree with DMCA or suing napster or whatever. I just think there actually exists another side to the argument.
They don't work, anyway. As a sysadmin I ocasionally check the logs for the filtering software (Of course, I go around it, don't tell my employer;) Every time I've seen someone try to find a porn site- he finds it. Sure, he gets any amount of blocked sites but, eventually, he gets one or two non-blocked ones. Also we continually have to adjust the policies to get around the false positives...
Slashdot Mirror
On the other hand, I think that "real" AI would most probably be created by "lesser" AIs. Which I would hope would dilute the human element. And we already have human "bugs" endangering life. Anybody up for a bit of human sharpshooting from a high tower?
Definitely an interesting subject, and one well worth thinking about, for unless there is something "magic" about humankind (highly unlikely, in my opinion), AI *is* coming.
Granted, this post could be considered an unlimited resource; but the time, effort, experience, whatever, that led and allowed smack_addict (and later lurker 786 ;) to create it is not. In other words, the number of posts by smack_addict is not an unlimited resource. Not to mention the number of posts by X moderated up to Y. What about the universe of good posts? I'm sure everybody can extrapolate to other creative endeavours.
Which is not to say I necesarily agree with DMCA or suing napster or whatever. I just think there actually exists another side to the argument.
They don't work, anyway. As a sysadmin I ocasionally check the logs for the filtering software (Of course, I go around it, don't tell my employer ;) Every time I've seen someone try to find a porn site- he finds it. Sure, he gets any amount of blocked sites but, eventually, he gets one or two non-blocked ones. Also we continually have to adjust the policies to get around the false positives...