It's no secret that robots are here and they are doing a lot of work. Humans and robots have been working alongside for over 35 years. The speculation has gone from intelligent machines (Ray Kurzweil in the 80s) to spiritual machines (Kurzweil in the last decade). The first human to be killed by a robot, Robert Williams, was decades ago - in 1979. Obviously that whole "Three Laws of Robotics" thing is a ways off. It's fine to say "A robot shall not harm a human or by inaction, allow harm to come to a human", but the sophistication to decide what's human and what isn't human would choke the biggest computers on earth today.
Allow me to introduce Heather Knight, Ph.D. student and owner of the (fantastically named) company, Marilyn Monrobot. Engineering magazine Design News had an article on her work that drew my attention to it. Her interests are in making robots "live" better with humans. Making them "Devilishly Charming Robots and Charismatic Machines." Robotics researchers talk of something called "the uncanny valley":
Part of her mission is to address the so-called "uncanny valley" -- a moniker used by roboticists to describe the phenomenon wherein humanoid robots give the creeps to real humans (which most of you probably are).There's a fine line between making robots that move with us, fit in with us in our homes, but that don't creep us out. I have seen some anthropomorphic robots that do creep me out a bit.
A robot shall not harm a human or by inaction, allow harm to come to a human", but the sophistication to decide what's human and what isn't human would choke the biggest computers on earth today.
ReplyDeleteIndeed, and then there's also the question of what constitutes "harm". Even the greatest of human minds have grappled with this one for millennia as well.
Robots and artificial intelligence are fascinating fields, but I think there needs to be much more understanding of the fundamental processes of thought and consciousness before we really start to see the cool stuff. Kurzweil describes some interesting concepts, and presents some great targets to aim for, but the complexities of realizing those concepts cannot, IMO, be overestimated. Making robots that act like humans is one thing, but making them think like humans is quite another. Come to think of it, I'm not so sure that would be a good thing. A machine can't be a slave (as we think of the word) but what about a sentient robot that is bound to one purpose only - would it be a slave?
On a lighter note, little robot humor (as told by the ventriloquist robot on one of my kids 321 Penguins DVDs)...
Two hydrogen atoms are walking down the road when one turns to the other and says, "I think I just lost an electron." The other one says, "Are you sure?" The first atoms answers, "Yes, I'm positive."
But then again, some of the work being done is absolutely fascinating - check out Al Fin on this type of stuff...
ReplyDeleteFembots. Just sayin'.
ReplyDeleteLeverAction, That story on the bio-compatible memristors at Al Fin was really cool - thanks for the link. After the initial "we're going to change everything" story in EE Times when the memristor was announced, I haven't heard a thing about them.
ReplyDelete