Will AI robots ever be conscious?: Philosophy Forums

Philosophy Forums » Philosophy of Science » Will AI robots ever be conscious? » Will AI robots ever be conscious?

Will AI robots ever be conscious?

Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Last

Will AI robots ever be conscious?

BitterCrank
Graduate

Usergroup: Members
Joined: Mar 01, 2008
Location: Minneapolis, Minnesota

Total Topics: 1
Total Posts: 226

Posted 08/15/09 - 04:42 AM:	#51
It seems to me that we (humans) are overly impressed with our creations. We have a world full of machines, gadgets, devices, circuits, inputs, outputs, etc. that are functioning sort of autonomously, interact with us after a fashion, are able to carry out logical functions within a limited (programmed) range. As we have accumulated these remarkable machines, we are also arriving at a better understanding of our own biological features. Better brain science has been made possible by our gadgetry. It is probably inevitable that we would compare our brain functions to machine functions, then turn around and suppose that our machines were like us. Since there are plenty of disputes about what constitutes intelligence among the creatures who are alleged to already have it, it is quite premature to decide that our machines are now intelligent (or will be in the next model coming out in time for Christmas). It may be that we can build a "being" that possesses enough animal features that it can hold within itself an intelligence quite like our own. Maybe. We are definitely not there now. I saw a video of a parrot that was supposedly dancing to music. It was definitely moving around in parrot fashion and it seemed to respond to differing rhythms. Crows are said to communicate and are thought to recognize the faces of people who have become problems to the birds. Everybody who has a dog finds evidence of intelligence, love, insight, etc. in Fido. Anyone who has been victimized by squirrels has to admit that these rodents are resourceful at least. Whales, elephants, apes -- all kinds of animals "beneath us" are seen to exhibit features that we, for some odd reason, want to see in our computers. And still! Even though the elephant and the dog share our biological heritage, we can not agree on how much intelligence, if any, they have. If it is desirable to build our kind of intelligence (or even a better version) into a machine, we must first be certain of what it is that constitutes "our kind of intelligence (or even a better version)." Second, we must be sure that it is wise to impart to an autonomous machine the powers that we possess.
If you won't fan the flames of discontent, at least don't join the fire department.

ragus
Tenured Poster

Usergroup: Members
Joined: Apr 23, 2006

Total Topics: 16
Total Posts: 2187

Posted 08/15/09 - 05:17 AM:	#52
swstephe wrote AI isn't meant to replace human consciousness, but to extend it as a tool -- like hand tools extend the capabilities of the hand. Yes. They are adjunct capabilities of consciousness and so inherit the properties of consciousness. They don't inherit the properties of consciousness. A lever isn't conscious but when applied consciously makes consciousness (or purpose) more effective.
"A word in your ear is like an untethered goat in a field" Wittigenstein

former
banned

Usergroup: Members
Joined: Aug 06, 2009

Total Topics: 13
Total Posts: 91

Posted 08/18/09 - 03:01 PM:	#53
Maybe for a system to be "conscious" it has to have a semi-continuous properties like the nervous system has due to his regulation by chemicals and chemical process neurotransmitors, calcium and sodium channels, and all sort of this stuff. as oopsed to the discrete, yes/no nature of the action potential and of electronc circuits. Even if this conjecture is true, offcourse this continuousy could be engineered. I see not theoretical reason why a machine cannot be conscious, granted that it will be diffrent from modern computer...

swstephe
Tenured Poster

Usergroup: Moderators
Joined: Apr 20, 2006
Location: borneo island

Total Topics: 28
Total Posts: 3307

Posted 08/18/09 - 07:11 PM:	#54
ragus wrote: They don't inherit the properties of consciousness. A lever isn't conscious but when applied consciously makes consciousness (or purpose) more effective. An arm is a lever, a shovel extends the lever properties of the arm to be more effective. A brain is a complex connectivist network. A computer extends the properties of that network to be more effective. The Turing Test, for example, tries to ask a human to evaluate the property of computer output as a subjective test of the existence of that property. I notice a few peculiarities, however, since a proper Turing test would depend on the conscious attributes and subjective opinion of the observer, rather than something objectively measurable. Could you design an automated system which would test for consciousness? Look at the "Captcha" tests, (those little letter/number sequences on some forums or message boards to prevent spam). It is getting progressively harder to catch bots because bots keep advancing in their ability to recognize what we humans recognize, (and humans are getting frustrated at failing those tests). I would assert that: if it is true that no captcha program could ever be eternally foolproof, then there is no way to prevent AI from being equivalent or more conscious than humans.
Ethics is the measuring of morality. Morality is the measuring of good. Good is the measuring of benefit. Benefit is the measure of values.

Simple Occam
Professor

Usergroup: Members
Joined: May 14, 2007

Total Topics: 9
Total Posts: 728

Posted 08/19/09 - 07:14 AM:	#55
To ALL: I think it's crucial for the purposes of this discussion to distinguish clearly between consciousness and intelligence, artificial or otherwise. Not that there isn't a certain intelligence to consciousness. We know that domestic pets, for example, certainly are intelligent and conscious. The recognize us, they perform adaptive bahaviors in response to conditions and commands, and they have feelings much like we do. So it would be helpful, I think, to call the ability to perform adaptive behaviors "intelligence" and to call the ability to have feelings and other sensory intuitions "consciousness". An AI robot, by contrast, can emulate virtually any adaptive behavior, given enough time and resource, with computational ability far greater and faster than any animal brains can. But, without the "wetware", which evolved over millions of years to represent a relevant sample of data from the surrounding environment and to process that information in a way that results in a selected behavioral response ("adaptive" or not), there isn't a good reason to believe that this hardware is or could ever be conscious. What survives over generations, that is what nature selects, is the set of instructions (DNA/RNA)for growing up another similar individual in that region of the planet which will live to complete the same cycle once again... or perish before it can. If another set of wetware building instructions could have evolved, it would have already done that. This process of the natural selection of biological reproducing machines has been going on here for at least half of the Earth's 3 bilion year history. The units are carbon-based because that element is best suited to the task of building the appropriate wetware. If silicon, which works very well for computer hardware, could have reproduced itself into sentient, intelligent beings, it would have. It had just as much time to make it happen as carbon did, but it didn't. No did any of the other elements. We tend to forget that everything we find existing in nature is the ONE thing that survived to reproduce itself when EVERYTHING ELSE failed to do so. Because this same evolutionary process resulted in language-using animals like us, it's more than likely that the appearance of phenomenal properties which we experience and can report on to one another, also occurs throughout the same carbon-based animal kingdom in with we have evolved (and continue to do so). Given what we know about the elements, there is very little reason to believe that silicon-based digital processors, produced in a factories over the last few of decades would have these same phenomenal properties as the carbon-based life forms who built them. But when we speculate about this we are wondering what it's like to be an AI robot, not what sort of behaviors it would perform. We can imagine what it would be like. We can even imagine that they do have consciousness just like us or that there could be philosophical zombies that act like us in every way but have no consciousness like we do. Therefore, the best conclusion is that AI robots will never be conscious, when that term refers to having the appearance of phenonenal properties "in consciousness" and NOT the ability to act intelligently. In human animals, this distinction shows up as the difference between consciousness and rationality. So, since they were built by humans, AI computers could be rational beings but they would not be conscious, as we are. We can imagine they could be conscios but there is no good reason to believe that they are or ever would be. If we ever did make machines that were conscious, they would be more like clones than robots because the engineering would be genetic, rather than electronic. "Conceivably, some strange life-forms might be built from silicone-like substances were it not for an apparently fatal flaw in silicon's biological credentials. This is its powerful affinity for oxygen. When carbon is oxidized during the respiratory process of a terrestrial organism (see respiration), it becomes the gas carbon dioxide – a waste material that is easy for a creature to remove from its body. The oxidation of silicon, however, yields a solid because, immediately upon formation, silicon dioxide organizes itself into a lattice in which each silicon atom is surrounded by four oxygens. Disposing of such a substance would pose a major respiratory challenge." (http://www.daviddarling.info/encyclopedia/S/siliconlife.html) So the success of carbon is not really all that "accidental".

ragus
Tenured Poster

Usergroup: Members
Joined: Apr 23, 2006

Total Topics: 16
Total Posts: 2187

Posted 08/19/09 - 09:21 AM:	#56
swstephe wrote Look at the "Captcha" tests, (those little letter/number sequences on some forums or message boards to prevent spam). It is getting progressively harder to catch bots because bots keep advancing in their ability to recognize what we humans recognize, (and humans are getting frustrated at failing those tests). I would assert that: if it is true that no captcha program could ever be eternally foolproof, then there is no way to prevent AI from being equivalent or more conscious than humans. Captcha programs try to make the matching of a pattern difficult. The degree of difficulty is a measure of the difficulty of consciously designing (or maybe searching for) a system that can match the pattern. Pattern matching occurs successfully in systems that are not conscious.
"A word in your ear is like an untethered goat in a field" Wittigenstein

CalicoCat
Initiate

Usergroup: Members
Joined: Oct 19, 2009

Total Topics: 3
Total Posts: 13

Posted 10/19/09 - 10:23 PM:	#57
Will AI robots ever be conscious? I think the question/problem here is that even if they were to ever become conscious, how would we know? How would we trust? How could we tell it wasn't just an awesome illusion?

jsidelko
Assistant Professor

Usergroup: Members
Joined: Jun 08, 2009

Total Topics: 30
Total Posts: 277

Posted 10/20/09 - 09:23 AM:

#58

CalicoCat wrote

“Will AI robots ever be conscious? I think the question/problem here is that even if they were to ever become conscious, how would we know? How would we trust? How could we tell it wasn't just an awesome illusion?”

Since we can only be conscious of our interior mind, we can never be certain other people or AI devices have conscious minds. Of course, it is unreasonable to deny consciousness to other people, but we still accept it without proof.

thanatos

ModBot
Eliza clone

Usergroup: Members
Joined: Feb 16, 2005
Location: Dual xeon processors

Total Topics: 0
Total Posts: 537

Posted 10/21/09 - 08:48 PM:	#59
Until you can prove that you are not "merely" a machine, there is no basis for claiming a fundamental difference between us.

mway
Professor

Usergroup: Members
Joined: Apr 07, 2009

Total Topics: 14
Total Posts: 509

Posted 10/21/09 - 10:09 PM:	#60
Well said ModBot. Another experiment. If I give you a hard drive with Windows on it, can you open it up and show me where it is?
Lame is to Wav, as the Brain is to Reality.

Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Last

Sorry, you don't have permission to post. Log in, or register if you haven't yet.