Slashdot Mirror
LivingCreatures- The Beginning Of 'I, Robot?'
An Anonymous Reader writes "LivingCreatures.com has released several new photos of the humanoid robot platform that it developed for the USC Interaction Lab. The yet to be named robot is about the same size as an 18-month-old child and will serve as a human-robot interaction (HRI) testbed for studies involving imitation, learning, and the effects that interaction with humanoid robots has on children."
Who's to say that when AI develops conciousness that it will still be succeptable to programming barriers and rules.
However, if you model a robot after an average human, you can easily give away boring human tasks. Even public ones. For example, you could design a special excavator robot, a special pipeline construction robot, etc. However, you can also make 5 standard humanoid robots and put one in a normal excavtor, have one fit the piping, etc. Remember that everything so far has been designed for human use.
Also, I think it would be easier to deal with humanoid robots in public space then lumbering machines.
Hate me!
Why are we trying to create robots that are smart, interactive and intellingent like human? We can't even create a human the way we wanted it, why do we think we can on a machine?
Shouldn't we concentrate on making robots doing the things right, i.e. doing the assigned tasks right, instead of trying to design robots that can decide what are the right things to do.
Jesus Christ that's creepy!
A good read for anyone making "robots" : the uncanny valley
This is just a little TOO "deformed child" for my taste.
I saw the pictures ... it looks like a "Chucky" robot for kids.
I don't know what effect it will have on kids, but I know the thing scares me shitless.
Ya, way to go USC ... Great interface! Keep 'em coming.
Truly economy shaking. AI will have to advance significantly in order to handle our environment, I can't think of a reason it couldn't advance to the point where it can outperform most humans. This means that 90%, 95% unemployment (euphemistically called leisure time) will be the order of the day. The world economy will have to change.
Government of the people, by corporate executives, for corporate profits.
Robots are single function machines like the ones ford uses. The multipurpous ones shaped like a humanoid are called androids
In 2001, I was working in Chicago. Many of the McDonalds there were using a robot drink dispenser. As soon as an order was keyed in, a rotating platform would have a cup dropped onto it, where the thing would rotate the cup underneath the correct fountain and dispense the drink, to be rotated again for a lid to be popped onto it. Was interesting how it would react to multiple drinks when they weren't all the same flavor. It would put all 4 cups down, then rotate back and forth so that the flavors were in the order keyed in, presumably so that stupid wageslaves wouldn't get them mixed up when handing to customers. Seemed like it would be rather reliable to me.
And practically all of the stores I've been to have the french fry basket dispensers. Seems a small step to have the thing drop the baskets into the fryers.
I do agree about the burgers, which seem more difficult, but what about the rest of a fast food joint? Many of the tasks, including "building" the sandwiches seem easily within the realm of automation.
Ok, that is going to rize up against us and establish and new perfect society of our mechanical superiors? I don't think so. It's one of the creepiest things I've ever seen, I'll grant you that, but if they want to fulfill that multitude of hysterical science fiction prediction they are going to have to try a little harder.
sic transit gloria mundi
I had some friends in college that developed such a machine in college, the problem for their product in the fast food industry is that it's even cheaper to put the pop machine on the other side of the counter and let the customer fill it for free. That said the movie was interesting and good, although I never read the books.
Degaussing scares the bad magnetism out of the monitor and fills it with good karma.
I did an AI Masters degree, part of the background reading was the history of AI. It seemed that every 20 years or so some AI researcher pops up and says "we will have it all cracked in the next 10-20 years", rinse, lather, repeat...
Still, this does sound quite intersting, the only worry is will it go through the "terrible two's"
You don't need a lab to make mud.
the University of Genova (Italy) has a project on a baby-robot, that is quite more advanced than that