First  Prev  1  2  Next  Last
Artificial intelligence
Posted 4/7/09

Cuddlebuns wrote:


khryzalids wrote:

well.. without emotions or free will machines wont move...


Unless you've live in a cave, I'm pretty sure you've seen tons of machines that move that don't have emotions or free will. Ever heard of cars?


And cars move because?
78167 cr points
Send Message: Send PM GB Post
29 / M / Japan
Offline
Posted 4/7/09

ShroomInferno wrote:


Cuddlebuns wrote:

Unless you've live in a cave, I'm pretty sure you've seen tons of machines that move that don't have emotions or free will. Ever heard of cars?


And cars move because?


Because a person wills it, right?


Well, I think a better example would not be limited to simply "motion". The basic foundation of any AI system is that it requires some form of input before it can perform a certain output (which is somewhat similar to how human sensorium functions through affect and effect), and one good example is probably the thermoregulator you find in an air conditioning unit.

It's a very simple AI system that measures the temperature within the room as it conditions it, making sure that the temperature remains at the prescribed temperature. To perform this, it may turn on or off the rotary fans inside to increase or decrease airflow through the unit, etc.

But of course, you can always argue on the grounds that the aircon had to be turned on first before it could perform such. I don't think the point of any AI system is to be sentient in any way, rather, the point of any AI system is to perform "intelligent calculations" as a thinking person would so much so that it requires less and less input from a human source.


I'd honestly throw AI as a sentient being out the window due to the sheer complexity of human emotions or even free will... simply because people still argue today what exactly those two things are... but that's something to discuss in another thread, besides this one.
5229 cr points
Send Message: Send PM GB Post
24 / M / Mammago Garage, Y...
Offline
Posted 4/7/09 , edited 4/7/09

ShroomInferno wrote:


Cuddlebuns wrote:


khryzalids wrote:

well.. without emotions or free will machines wont move...


Unless you've live in a cave, I'm pretty sure you've seen tons of machines that move that don't have emotions or free will. Ever heard of cars?


And cars move because?


Because someone or something presses on the gas pedal and makes it move. If you're trying to say that only something with free will and emotion can move on it's own, then that doesn't really make sense, because even a machine with free will and emotions would need someone to turn it on. And there's plenty of organisms with no free will or emotion that can move as well.

Or you can just ignore me and read what edsmac posted.
9980 cr points
Send Message: Send PM GB Post
116 / M / Earth
Offline
Posted 4/7/09

Cuddlebuns wrote:


khryzalids wrote:

well.. without emotions or free will machines wont move...


Unless you've live in a cave, I'm pretty sure you've seen tons of machines that move that don't have emotions or free will. Ever heard of cars?


what do you mean? ofc a car will move if u leave it in a very steep slope..........
Posted 4/9/09 , edited 4/9/09

edsamac wrote:


ShroomInferno wrote:


Cuddlebuns wrote:

Unless you've live in a cave, I'm pretty sure you've seen tons of machines that move that don't have emotions or free will. Ever heard of cars?


And cars move because?




Mostly agreed. Especially the part in bold.

_________________________________

As I can see the OP's stance on this subject is that the AI robot is some sort of machine with superhuman intelligence.

That's what the average population is generally assuming - that computers, or some kind of machine will eventually achieve superhuman intelligence and this form would be regarded as an AI.

But question is: Is there such a thing such as superhuman intelligence at all?

We all agree that there are various levels of intelligence across this world's biodiversity which are widely aknowledged. (ie. Bacteria is less intelligent than most insects, mammals more intelligent than reptiles, apes more intelligent than most mammals, and most humans more intelligent than apes, etc. - disregarding the existing exceptions) - there's always an animal that's notably more intelligent than the other - fundamentally superior.
The emotional aspect in this whole issue isn't something that we should underestimate. I think we all can agree that most intelligent species do have the ability of strong emotional attachment which, for example, can't be found in flies.

And I'm sure that we can all agree that human beings doubtlessly reside in the upper echelons of the intelligence spectrum on Earth. They can solve problems which, for example, a dog or horse couldn't solve regardless of the time permitted. - - they most probably could never develop the necessary mental ability to become the Earth's ruling species -- this all looks like certain species just can't surpass a certain level of performance(neurologically incapable?).

Point is, we can solve things which most species can't, but even we do have limits.

Ever heard of the turing equivalence? Well, first of all let me explain the meaning of turing complete. It's a way of mathematically stating certain levels of computational ability = Turing complete -- meaning a computer that is capable of solving all the problems that are named turing complete, or at the least simulate them -- which in turn can solve the problems.
From Mac to Windows, all of those systems can technically solve all the problems that the other systems can, the only difference is in speed - some are faster and some are slower, but in the end they all solve the problems just as good. They're turing equivalent. - to be turing equivalent suffice to have the ability to simulate or be simulated by all other turing complete computers.

Question is: Are we turing complete? Can we solve problems that those so-called turing complete computers can, or can we at least simulate them?
Well, obviously we can when we're thoroughly motivated. When we invented those computers we had to simulate them(by drawing for example).
Now, having previously mentioned turing equivalence -- are we turing equivalent? Can a turing complete computer simulate us? Theoretically a computer can simulate a human brain, and therefore compute and solve any problem that the human brain can solve; we're presumably turing equivalent.

It's true though that many computers can do things which we can't. They're unbeatable in chess, for example. De facto, they're already smarter than us in some domains. But does that mean that they are superhuman? Well, no. We still can simulate them, right? They just beat us by far when it comes to speed.

Now, if we ever got to create an AI with intelligence that is equivalent to ours, and it supposedly could create more AIs it could never design an AI that surpasses its own because of the turing equivalence. Right?

Well, we could argue now that speed is indeed a huge advantage and could be considered as one of the traits of being more intelligent thus surpassing us by far.

So if it were ever to happen that we'd create an AI that has all of our capabilities(simulations of emotions included), the fact that he can process all the data a lot faster he'll be able to solve the issue a lot faster than us. It could be regarded as potentially dangerous for ourselves. An AI capable of human emotion even if just simulated could do some disastrous things. However I doubt that he'll ever be able to feel what we feel, yet he'll be capable of simulating them, but not able to understand WHY and therefore probably not able of compassion. O_o







10513 cr points
Send Message: Send PM GB Post
27 / M / In your room stea...
Offline
Posted 4/9/09 , edited 4/9/09

Cuddlebuns wrote:

Because someone or something presses on the gas pedal and makes it move. If you're trying to say that only something with free will and emotion can move on it's own, then that doesn't really make sense, because even a machine with free will and emotions would need someone to turn it on. And there's plenty of organisms with no free will or emotion that can move as well.

Or you can just ignore me and read what edsmac posted.


Exactly, a mobile robot can move on its own, if it has basic decision making capabilties and wheels/treads, through basic programming you cam make a machine move on its own, and it has no free will or emotion.
18663 cr points
Send Message: Send PM GB Post
36 / M / Small Wooded town...
Offline
Posted 4/9/09 , edited 4/9/09

Allhailodin wrote:

Exactly, a mobile robot can move on its own, if it has basic decision making capabilties and wheels/treads, through basic programming you cam make a machine move on its own, and it has no free will or emotion.


'there are programs out there that allow a robot to think. Programs created to let a robot deside on the best way to get to point a from point b.
There are also programs that allow a robot to have a basic understanding of emotions.
10513 cr points
Send Message: Send PM GB Post
27 / M / In your room stea...
Offline
Posted 4/9/09 , edited 4/9/09

Darkphoenix3450 wrote:

'there are programs out there that allow a robot to think. Programs created to let a robot deside on the best way to get to point a from point b.
There are also programs that allow a robot to have a basic understanding of emotions.


yeah


the else-if handler is good at doing that (decision making)

if (obstacle = 1)
go_around()
else
keep_going()


it's obviously not that simple, because you would need to determine where the obstacle is ( in front of you, to the left of you, behind you, off to the right ) then need to figure out how to go around it, and so on, but you can do all of that in a program, and its not giving the robot any ability to think on its own, it just executes code as it sees it, but basic decision making is done with the else-if or switch handlers, easy to do.


and emotions are harder to do because you can't simply go


switch (emotion)
case 'happy' :
smile_cheerfully();
break;

case 'mad' :
beat_nearest_person();
break;

case 'sad' :
cry_and_act_depressed();
break;

and so on, not that simple, you can program how one would act when feeling a certain emotion, then simply make the code call the handler related to the emotion, but you can't really program a feeling into something. like you could program a machine to react in a certain way to something, like you could program a robot to yell and cuss at you if you punched it, but you cant actually make it get mad.
Posted 4/10/09
Yeah, I agree... it would probably destroy us.
Posted 4/10/09
BUT FUCKKKK IT
5348 cr points
Send Message: Send PM GB Post
29 / M / England
Offline
Posted 4/10/09
Pools closed due to aids.
First  Prev  1  2  Next  Last
You must be logged in to post.