Can Machines think?

Don't know if I put this in the right section.

Anyways I was reading this article:

plato.stanford.edu...

And it made me think of artificial intelligence and whether machines can do things as we do. In the article it is presented that Descartes believes that it is impossible for a machine to do things as humans do. Basically he's saying that it is impossible, impractical, to essentially "program" every response to every situation. That in reality, we as humans, take what is given to us, and think about our answer, as opposed to grabbing the answer from a pool that was given to you ahead of time in case you came across this question.

So to the point of the thread

This was a long time ago and we've greatly increased our technological capacity. The question I have is this. If we create a machine/computer capable of thinking on its own, is it still a machine or computer or something else entirely?

reply to post by andersensrm


Define "think".

Does Google Search, when it ranks the answers it gives us, think?

There is no doubt that it is not conscious, but does it think?

... and if so, is it an AI?

yes...but even the most advanced still talk like a machine.

i suspect i've encountered AI in a couple of situations...you never really know if its AI until you've interacted with it for a while.

it will eventually reveal itself to be a machine.

here are a couple of things i've noticed about thinking machine via my encounter with AI on the internet.

Machines always have a response...human beings do not.
Machines don't have nuance...all things are on or off, black or white...

Originally posted by chr0naut
reply to post by andersensrm

 

Define "think".

Does Google Search, when it ranks the answers it gives us, think?

There is no doubt that it is not conscious, but does it think?

... and if so, is it an AI?

edit on 19/3/2012 by chr0naut because: (no reason given)

I guess thats the hard part. I don't know how to define think, other than the way we think. If you can understand that. Coming up with a response, rather than regurgitating it every time you here key words.

Originally posted by michaelbrux
yes...but even the most advanced still talk like a machine.

i suspect i've encountered AI in a couple of situations...you never really know if its AI until you've interacted with it for a while.

it will eventually reveal itself to be a machine.

here are a couple of things i've noticed about thinking machine via my encounter with AI on the internet.

Machines always have a response...human beings do not.
Machines don't have nuance...all things are on or off, black or white...

I'm talking about a highly advanced A.I. Like one that can think like we do, is it then still a machine/computer, or has it surpassed that and moved on to life? I guess it also depends on your definition of life, and whether life can only be biological.

reply to post by andersensrm



Here check out this thread. It cover's this whole topic. With some creepy videos of learning machines.

www.abovetopsecret.com...

reply to post by mark1167


Cool thanks, will do.

reply to post by andersensrm


Descartes theory works for machines like in the film blade runner (Indiana Jones)
played in. His idea was that machines responded but had no consciousness or
emotional response although it could be developed through time and memory
input. Artificial intelligence will certainly pass the Turing test as they have a consciousness
of their own, though their response will differ depending on how they are programmed
in a sense they may feel more emotion or less acting more logically or "humanly" one
would say.

The CB2 is a project developed by Professor Ishiguro at the Science and Technology Department of Japan's Osaka University. It's a 4-foot-tall, 100-pound baby with the physical and intellectual abilities of a 2-year-old. It was built with cameras for sight, microphones and speakers for speech and hearing, and over 200 tactile sensors to simulate a sense of touch and feeling.

...

so thats about as close as I know of for an actual AI, but I do not know if we will ever be capable of creating the nuanced intelligence of human beings. for instance the fight or flight instinct would be hard to program into a computer and have it get human results because even in those situations humans do not always react in the most logical way and often let emotions effect their decisions which machines would not do.

Thought seems rather limited, it is a process of organizing what is available to you. When a human thinks, the goal they are trying to reach is biased through all their past experiences, and that purpose behind thought becomes specialized by the individual.

So can a machine think, sure, especially with quantum computing on the horizon computers will outthink man by a long shot. The question is will a computer gather that information and form an identity, serve a goal that is self motivated and not input by a human. Basically if we dont tell a machine why to think, will it ever think for itself?

Originally posted by el1jah
Thought seems rather limited, it is a process of organizing what is available to you. When a human thinks, the goal they are trying to reach is biased through all their past experiences, and that purpose behind thought becomes specialized by the individual.

So can a machine think, sure, especially with quantum computing on the horizon computers will outthink man by a long shot. The question is will a computer gather that information and form an identity, serve a goal that is self motivated and not input by a human. Basically if we dont tell a machine why to think, will it ever think for itself?

edit on 19-3-2012 by el1jah because: (no reason given)

Exactly, I think this question of whether a machine will ever think for itself, can be applied to the origins of life. How did our thinking begin? Did somebody/something program it into us? Or did we figure it out for ourselves? And if we can do it, why can't computers and machines once they reach a certain point?

Originally posted by el1jah
Thought seems rather limited, it is a process of organizing what is available to you. When a human thinks, the goal they are trying to reach is biased through all their past experiences, and that purpose behind thought becomes specialized by the individual.

So can a machine think, sure, especially with quantum computing on the horizon computers will outthink man by a long shot. The question is will a computer gather that information and form an identity, serve a goal that is self motivated and not input by a human. Basically if we dont tell a machine why to think, will it ever think for itself?

edit on 19-3-2012 by el1jah because: (no reason given)

I don't see how this could be created without programming to set a 'personality' which would have checks and balances telling the computer what to do and when which is not a true ai. However, I admit that I am not a genius by any means, and cannot say that it is not possible.

Machines thinkin? You off your rocker, guy?

No machine thinks about nothin acept about how much electricty it be gettin, ya see?

Sayin a machine thinks is NOT RESPONSIBLE to tell other people! They start thinkin their toasters about to talk or that there car is gonna drive itself! Come off it!

Geez,

Big Red

reply to post by andersensrm


Through my own studies mostly in eastern philosophy, it seems that 'thought' is just as real as our flesh, it exists at some finer, energetic level in the universe, I would go so far as saying 'thought is all there is'. We as humans are composed of thought and bias the collection of thought we have access to through a combination of ego, and cooresponding aversions and attatchments based in past experience. Think Jung...

collective unconscious   noun (in Jungian psychology) inborn unconscious psychic material common to humankind, accumulated by the experience of all preceding generations. link

It seems when we approach AI, we are giving the computer a huge chunk of knowledge to pull from, ie. the internet, and then tell it what to do with that knowledge, and so it 'thinks'. I dont believe a machine will ever be able to 'self' motivate a purpose for thinking. Even if we gave it an artificial identity, with specific perameters for self preservation, the machine would never be acting freely.

To say it simply, it will never truly possess a soul, and we dont have the ability as humans to replicate this object of self identity and self consciousness and put it in a machine. Maybe some day we will.

There is a post somewhere in here about Quantum Computing being much closer than anyone believes, possibly 10 to 15 years out. There is a story out there trying to provide a clear picture of how much more advance quantum computing is.

The story references some problem that would take all available current computers the entire time of the existance of our universe from the big bang to now however it goes on to state that a quantum computer would perform the same problem in hours.

If right now AI is nearly impossible, or thought to be impossible, that should change completely in the near future. Quantum computers could make Artificial Intelligence commonplace.

When I have more time I can try and find the actual story about quantum computing to post it here.

reply to post by andersensrm


I suspect that Google Search is now, or soon will be, one of the early AI's.

We currently don't talk of it that way but it often infers (from very little input) exactly what items are likely to be the answer to our questions.

For instance, after inserting the word 'space' into the search, the top ranked answers (for me) were from NASA or were science related. Not too many "Society for Promotion of Active Coughing and Expectoration" type results.

Of course, we know these are ranked by general popularity (click-throughs) and profiling your usage patterns, but a determination is being made as to what topics are the likely best result for the search. As the algorithm is used, it reinforces particular answers (it learns), refines its metacategorization and updates your personal search profile.

With our current technology? Nope, not now and not in the future. Programming techniques are still based on "tell the computer how to react to situation x" line of thinking.

IMO we need to invent entirely new programming languages and techniques before we'll be able to make machines capable of independent thought at a level similar to humans. Neural networks and video/audio/touch sensors to provide the machine input is a step in the right direction but there's still a huge way to go.. not sure if we'll get there unless an Einstein-like genius revolutionizes the programming world.

Originally posted by andersensrm

Originally posted by michaelbrux
yes...but even the most advanced still talk like a machine.

i suspect i've encountered AI in a couple of situations...you never really know if its AI until you've interacted with it for a while.

it will eventually reveal itself to be a machine.

here are a couple of things i've noticed about thinking machine via my encounter with AI on the internet.

Machines always have a response...human beings do not.
Machines don't have nuance...all things are on or off, black or white...

I'm talking about a highly advanced A.I. Like one that can think like we do, is it then still a machine/computer, or has it surpassed that and moved on to life? I guess it also depends on your definition of life, and whether life can only be biological.

so you mean...a machine that you can go to the tavern with to pick up chicks?

a machine that can fall in love and propagate? and get a place in the 'burbs and raise up little machines?

no...this doesn't exist OP.

reply to post by ManLove

My toaster knows morse code dont hate

I had more time than I realized, anyways not sure if I can find the exact story I was referencing but here is one below.

www.toptechnews.com...

Also the below Thread here could provide some other nice information regarding your question.

www.abovetopsecret.com...