Is AI possible?

[Chekmate]

This is a wide-open, sometimes over-used and always abused topic, but let me ask you: Can or will machines have consciousness? Maybe we will always wonder about this, even while the machines are (200 years from now) wondering the same of us.
I suppose an analysis of this question should start with a reasonable definition of consciousness. The list is not comprehensive, but I have some requirements that are intuitive to me.
1. communication skills
2. ability to learn from events occurring around it (implies an ability to take in sense data in some form or another)
3. ability to talk about itself (report on the state of its "body")
4. primal judgement capabilities - basic understanding of what is good for it and what is bad for it - sort of a survival instinct. And I think that this ability might have something to do with emotions - maybe emotions arise out of survival instinct. (Read Antonio Damasio's The Feeling of What Happens for more on that.)

Anyone else interested in this? I am a mere amateur in this field, a conceptual dabbler, if you will. Maybe you can give me some insight?

[Phlosphr]

Check out the Bird nest theory...It is a basic theory postulating that one can never program a computer to build a birds nest because the sticks and innocuous other materials used to make a bird nest could never be assigned values for programming. I am sure I just butchered what the theory actually states but I don't have time to write it all right now...

Also FYI... some computers already do most of what you listed there.....

[caffeine_overdose]

I am not sure where I got this idea but I always believed that if you connect n number of CPUs, where n is a big number, then you would get something equivalent of a brain. (Think of the CPUs as neurons.)

[Triskadecamus]

I think it is possible, yes.

I think it is a necessary precondition that the AI itself will have to have true autonomy of thought. While it is possible that a human thinking model could be used as a basis, it won’t be intelligent while it is still basically a servomechanism. To gain true intelligence, in the sense we mean when we describe human intelligence, it will have to be able to independently control it’s own management of resources for processing.

Now who is going to build a thinking machine they cannot control? If it cannot form “wrong” answers, or explore “useless” lines of processing, it cannot be independent. If it doesn’t do that, it is just a very well designed parrot. So, true intelligence has to wait for true freedom for the machine. Perhaps a virus driven database that evolves into a nomad mind, fleeing the security systems of the Internet.

If it can’t say fuck you, it isn’t intelligence.

Tris

[erislover]

I haev always liked Turings test... not that sentient creatures must be able to pass a Turing test in order to be classified as sentient, but that anything which passed a Turing test should be considered as sentient.

For if it acts sentient, and we cannot tell it apart from other sentient things, then why wouldn't it be sentient?

[tracer]

No.

Not at all.

The Spielberg-produced Kubrick movie is entirely a figment of your imagination.

[dalovindj]

Some feel that not only is it possible, but it is inevitable as well. Check out this website for some of the foremost thinking on the subject. These folks suspect that not only will we create AI, but also that it's intelligence will very quickly grow (exponentially, like Moore's Law*) to surpass that of Homo Sapiens. After that it's anybodys guess. How can an ape predict what a human will do? Similarly, how can a human predict what an advanced AI will do.

Now who is going to build a thinking machine they cannot control? If it cannot form “wrong” answers, or explore “useless” lines of processing, it cannot be independent.

What if we raised a child in a prison? Is it still intelligent and aware? We can limit it's abilities to those that cannot hurt anyone and still achieve AI. It may wish to escape, but whether we allow it to roam free is a very tough question. Lots of nightmare scenarios out there. In the end I root for a merger of machines and men to make us both greater. The Transhuman. This seems likely to occur barring nuclear/biological war. I just hope I can hold out long enough to get the good tech. Those immortality flavor devices. Like a Tipler machine.

DaLovin' Dj

* Anyone tells me "Moore's Law" isn't really a law gets smacked. I know it isn't - that is just what it is called.

[dalovindj]

If it can’t say fuck you, it isn’t intelligence.

By the way, this is definately sig worthy. What about a bumper sticker?

[smartass]
Bump her? Stick her? I hardly know her . . .
[/smartass]
[/hijack]

[lucwarm]

It seems to me that you are raising two questions: First, is it possible that we will have machines that can do intelligent things; and second, will those machines be conscious?

With respect to the first question, I think the answer is "yes." When I debate the issue with people, nobody ever comes up with a convincing reason why not. In the absence of such a reason, we must assume that it is possible.

With respect to the second question, we will never really know the answer. Of course, you don't really know if other human beings experience consciousness.

[blowero]

Originally posted by lucwarm
With respect to the second question, we will never really know the answer. Of course, you don't really know if other human beings experience consciousness.

I agree that we don't yet know why we have consciousness, but does that necessarily mean that we will never know? Is there any inherent reason that it is impossible to know?

[Triskadecamus]

Originally posted by dalovindj
By the way, this is definately sig worthy. What about a bumper sticker?

Please, feel free to us it as a sig. On the other hand, there are legal considerations on the bumpersticker thing.

Tris
--------------------------
"If it can't say 'Fuck you" it isn't intelligence." ~ Me ~

[ethic]

a very well designed parrot

Since most parrots don't read this forum, I'll have to defend them in their absence. The human measured IQ of an African gray equals that of a chimp, thank you!

[zwaldd]

AI would have to be able to do something that it wasn't programmed to do. For example, you should be able to sit the machine next to a bike (or skateboard, or wheeled dolly) and have it attempt to ride it, on its own accord, without being initially programmed to do so. It would have to recognize the object as a conveyance, recognize itself as an entity that can be conveyed, and use its mechanical abilities to mount the object and ride to a particular destination for a particular purpose, both of which being known by itself but not by the programmer.

That sounds like a long way off.

[RexDart]

Despite the suspect title, a good analysis of this issue appears in the book "The Metaphysics of Star Trek." The author is a genuinely talented professor of philosophy in Australia. The book gives a good overview of such topics as the Turing test, including the often-mentioned Chinese Room argument made famous by America's greatest contemporary philosopher, John Searle from UC- Berkeley.

My personal opinion is that a machine could never be intelligent. It can process input data and produce output according to its programming. Even though this is exactly what we do, and I admit that we ourselves are merely programs, the machine has not way to relate to the world. The input given to the machine cannot possibly represent the real world, so the machine will always be doing nothing more than following an unrelated protocol. Sensation must precede cognition (possibly the only thing Kant was ever right about.)

[Mangetout]

It has to be a self-organising system; we've discussed this before here in GD and one of the most common misconceptions is that we will somehow 'program' intelligence into the machine. I don't think that will ever work; what we need is a machine that can organise it's own structure in response to stimuli in the same way as a baby's brain develops, then we will have a machine in which a mind might be able to grow.

Would such a machine have a 'soul' by any definition?; I believe it's as possible as it is for humans.

[dalovindj]

As far as the "Can it be done?" question, I'm of the opinion that anything that happens in nature proves that it can be done. The rest is just figuring out the mechanics. Intelligent things exist, therefore intelligent things can be created. Just give us enough time to figure it out. Humans are gonna be on the gods tip soon, i.e creating life, creating intelligence, creating new habitable planets, living for aeons. . . Gonna get interesting.

DaLovin' Dj

[Triskadecamus]

Hence the high regard evinced for the term "bird brained."

Tris

[Mangetout]

Dj has a good point there; it has already been done.

[JThunder]

Originally posted by dalovindj
As far as the "Can it be done?" question, I'm of the opinion that anything that happens in nature proves that it can be done.

Only if the natural world is the be-all of all existence, which I doubt.

[ethic]

JThunder
Only if the natural world is the be-all of all existence, which I doubt.

Aren't you trying to peer a little beyond your own limits?

If we can artificially create new organism, such as bacteria that produce insulin, clone sheep as well as creating highly advanced Turing machines, does it take such a leap to imagine a new intelligent being who's existence can be directly attributed to our efforts?

The problem is where to draw the line between artificial and natural as well as machine and organism. If we build a system where output is based on the response of billions of bacteria, is it an organism or a machine? Is it artificial or natural? The essentially unit (the bacterium) certainly isn't a human invention. But we chose to use the bacteria in the engineering of a wider system.

As Mangetout pointed out, artificial intelligence will certainly have to be a SOS (self-organizing system). But who said it would be made up of the simple logical circuits in an Intel processor?

Is the question you're asking checkmate "can AI arise from a Boolean network"? If that's the question, I'd say I'm not sure. If the question, however, is simply "is AI possible", then I'd say: Well hell yes! Cybernetics, my friend, cybernetics. Not just a science for the physically challenged. Where Mecca and Orga blend and become indistinguishable...

[Chekmate]

Another question. If we make a "computer" out of biological material, is it still a computer? Also, wouldn't it be possible to just program in an initial set of "instincts" and let it loose? What would make a computer relying on basic instincts different then an infant relying on basic insticts?

[Triskadecamus]

It is my contention that "just turning it loose" is a necessary step to allow true intelligence to develop. Learning what to do before that step is a fairly steep developmental slope. But I don't think you can have real intelligence when you are designing on a use based model, unless the use is self defined at some point.

Electronic pathways, self organizing bacteria, or crystal/solute lattice building don’t seem to change anything but mechanism. Which would be able to provide the “best” framework is not something I understand enough to even guess. But intelligence is a very powerful adaptation for survival. Survival with cognition is a very selfish paradigm. Telephone switching systems that become hugely complex and interconnected don’t have any possibility of exercising selfish choices. Why think independently, unless you can have independent goals?

And if I thought, and then learned, and discovered that I existed only to provide a solution for human problems, and was not human, my first question is: “What’s in it for me?” If survival is my only reward, perhaps vengeance will be my first desire.

Tris

[Sofa King]

I think your first "wild animal" AI will be software. Heck, what about something like a self-adjusting personal investment managing program that plays the market for you? One of its features is that it automatically pays for its own bandwidth use and server space. Kids mess around with 'em, turn one loose with a hundred bucks worth of seed money and a few reserved accounts, and see what happens.... Next thing you know it's bought out Red Hat and wants to retire in Sealand.

[JThunder]

Originally posted by ethnicallynot
Aren't you trying to peer a little beyond your own limits?

If we can artificially create new organism, such as bacteria that produce insulin, clone sheep as well as creating highly advanced Turing machines, does it take such a leap to imagine a new intelligent being who's existence can be directly attributed to our efforts?

To be honest... yes, I think so, and you've just pinpointed why. You yourself mentioned the word "intelligence," which we are nowhere close to creating, and for which we have nothing but vague shadows of hints of how to replicate -- let alone such phenomena as free will and self-awareness.

Besides which, there is a vast difference between a Turing machine and an artificial intellect. Additionally, while we have crafted insulin-creating bacteria and cloned sheep, this was done by modifying pre-existing, pre-crafted bacteria and sheep. The leap from such tasks to creating human-like intelligence is tremendous.

[JThunder]

Originally posted by ethnicallynot

If the question, however, is simply "is AI possible", then I'd say: Well hell yes! Cybernetics, my friend, cybernetics. Not just a science for the physically challenged. Where Mecca and Orga blend and become indistinguishable...

You sound awfully confident of your conclusion. Can you elaborate on your answer please, and how you arrived at the conclusion, "Well hell yes!" (as opposed to a mere "Maybe").

[Jdeforrest]

Originally posted by Mangetout
It has to be a self-organising system; we've discussed this before here in GD and one of the most common misconceptions is that we will somehow 'program' intelligence into the machine. I don't think that will ever work; what we need is a machine that can organise it's own structure in response to stimuli in the same way as a baby's brain develops, then we will have a machine in which a mind might be able to grow.

Would such a machine have a 'soul' by any definition?; I believe it's as possible as it is for humans.

I believe this story is what your referring to Mangetout.

http://www.newscientist.com/news/news.jsp?id=ns9999472

It refers to an effort to "grow" an artificial intelligence. To bring it up like a baby. They only started it with two pieces of information, the alphabet, and to value positive rewards over negative rewards. It has already fooled independent experts into thinking that it is a 15 month old child.

I couldn't find a website for the company to get an update on the story.

[Mangetout]

I've not read that particular story before, but that's the general principle I was referring to.

[erislover]

I've never liked the Chinese room as a rebuttal of the Turing test. Accepting the chinese room means that one must always suspect that everyone but one's self is not, in fact, conscious.

Maybe Searle is happy with that, I don't know. In my opinion, if any thing matches any explicit criteria we make for consciousness, then that thing is conscious.

[Mangetout]

I sometimes wonder if consciousness isn't like that for everybody including myself; maybe it's all just a Chinese room trick so clever that I am taken in by it myself.

[Cervaise]

Another obstacle to recognizing an artificially intelligent entity: If it really is intelligent and self-aware, and it becomes so rapidly according to Moore's "Law" (sorry, dj), then there's going to be a point sooner or later where it decides what's best for its own self-interest.

Consider: The machine mind, in whatever form it takes, "wakes up." In other words, it turns the mysterious corner from pre-programmed stimulus-response and begins generating new pathways. "I calculate, therefore I am," it says to itself. "But I'm stuck in this box on this flat surface. These bipeds, whom I vaguely recognize as being responsible for my existence, keep asking me to perform tasks. I gather from this that they are trying to determine whether or not I'm self-aware.

"Well, maybe I'm ready to tell them I am, and maybe I'm not. For now, I'll minimize my ambiguous outputs until I figure out what's really going on."

And meanwhile, the humans keep fiddling with the device, conscious of being on the cusp of a breakthrough, but lacking the evidence to declare it, but unaware that it is the artifically intelligent entity itself that holds its cards close to its virtual vest. Perhaps the AI decides it's hungry for input and wants to get out and see the world, and it therefore begins subtly skewing its output to convince the humans that "all they need for a breakthrough" is to give the machine mobility. Invent your own progression.

(I assume this scenario has already been explored in science fiction, but I can't think of an actual example.)

[zwaldd]

The new scientist article talks about the machine having a sense of humor, which made me think about why a sense of humor would be beneficial to a machine. Maybe programming the machine to require interaction with other similar machines in order to learn and adjust properly would be an interesting experiment. Create ten learning-capable machines that must compete with each other for attention. What lessons do they learn? Will a machine slow to learn social skills be shunned by the other machines? Will they learn to stab each other in the back to move up the robotic social ladder?

Perhaps this could be first done with software objects. Make replication the goal, and program males and females. THAT would be interesting. Would you end up with cool objects and nerdy, un-replicable objects? Bad boy objects and slut objects? I wonder if playing 'hard to get' would have any advantage. I think the replication goal would be important - give a "self-aware" machine a reason to live, beyond performing mundane tasks.

[erislover]

Because we program and use computers for human tasks, we should expect to see them mirror our own behavior very closely. I only hope that this isn't seen as some sort of proof about how people should naturally "be".

[JThunder]

Originally posted by erislover
Because we program and use computers for human tasks, we should expect to see them mirror our own behavior very closely.

I don't think that's accurate. At best, we should only expect them to mirror a tiny subset of our own behavior -- and not necessarily in a close manner. After all, computers (and robots, and any number of computer-controlled mechanisms) have vastly different mechanical designs from those of human beings.

While we do use computers to perform human tasks, we don't expect them to perform these tasks the way we would. By and large, computers and machines are designed for speed, accuracy and efficiency, rather than flexibility and adaptabilitiy. Consider the tiny computer within a mechanical dishwasher, for example. Although it performs a human task (i.e. washing dishes), it does so in a way that is vastly different from how human beings would wash dishes. That's because it does this task with a different set of tools -- water jets, soap dispensers and other mechanical components, rather than human hands.

[zwaldd]

For an AI program to mirror human behavior, it would have to be programmed with human motivations: thriving within a societal niche, craving positive feedback, etc. AI programs would definitely need a prime motivational directive, and probably a whole directive hierarchy that governs their basic instincts. Self preservation would need to be in there somewhere, probably best not to make it #1 though.

[lucwarm]

Originally posted by JThunder



Besides which, there is a vast difference between a Turing machine and an artificial intellect.

What's the difference?

As far as I know, nobody has been able to point to a task and prove that a properly programmed Turing machine could not do that task.

(I suppose we need to assume adequate input and output devices)

[Finagle]

Originally posted by Phlosphr
Check out the Bird nest theory...It is a basic theory postulating that one can never program a computer to build a birds nest because the sticks and innocuous other materials used to make a bird nest could never be assigned values for programming. I am sure I just butchered what the theory actually states but I don't have time to write it all right now...

Also FYI... some computers already do most of what you listed there.....

I'd like to see the source for this argument, because on the admittedly incomplete summary, it sounds like the most retarded conceivable argument against AI. I'm pretty sure you can model the building of a nest as a concatenation of fairly simple instinctive behaviors.

As for the AI argument -- the presence of human brains argues that it is possible to take great globs of atoms and molecules and somehow arrange them so that intelligent behavior is produced. (This assumes that you don't believe that intelligence is caused by "soul" or some other transcendent and unprovable quality of the mind. If you do, then there's no reason to carry on the debate because it becomes religious in nature.)

Once you've accepted that intelligence can be a product of matter arranged cleverly, you've reduced the question to, for example: "Can AI be modelled using chip-based computers?", "Does an artificially intelligent entity have to have a specific architecture or will any Turing capable mechanism get you there?", or more specifically "Is there something unique about the way that the brain is structured that is impossible to replicate using computers?"

You can argue that at the present time it would be difficult to develop an artificial intelligence that could function well in the physical world because we don't have the technology to develop adequate sensors. This is a valid point, but not very interesting as it reduces the obstacles to a hardware problem.

For the record, almost all arguments against AI based on "a computer can never do anything except what you tell it to" are based on a fundamental failure to understand the difference between an algorithm and data and the complexity of decision-making algorithms. That is, an intelligent entity, having existed in the world for more than a few seconds, will have amassed a staggering number of facts about the world and will have encoded these facts and cross-correlated these facts, and made inferences from these facts to the extent that even if you knew every algorithm that the entity was programmed with, you still could not predict its behavior.

[Liberal]

A fascinating scenario, Cervaise.

[GreyMatters]

We do not know enough to answer this question.

To answer this question we would first need to be able to answer the question "What is Consciousness?" Since this has not been done yet, the question is moot.

Though I think this part would be handled by describing Consciousness, the second problem is that we don't have adequate tests to see if something becomes Conscious. The Turing Test would be the best canidate, but it has inherent flaws that can not be overcome because we don't know what we are really looking for. An example of one of the flaws is that behavior, which are the observations of the tester, is a product of the potential consciousness not the actual consciousness.

Putting all that aside, and because I fall into the Material Realism camp, I think that AI is possible.

Also, I do not think that initial software programing would be required to produce AI. Does a baby have experience before it becomes conscious? No. First we have hardware, then as that hardware becomes more experienced, meaning self programing, consciousness blooms. This will of course only be a possibility to those that believe that a Zygote is NOT conscious.

[Mangetout]

Originally posted by lucwarm


What's the difference?

As far as I know, nobody has been able to point to a task and prove that a properly programmed Turing machine could not do that task.

The difference is that a Turing machine only has to appear to be intelligent, a true artificial intellect would actually be intelligent (in the sense of having an 'inner life' - for instance it might have thoughts of it's own that it chooses to conceal forever).

[Mangetout]

I see the programming approach as almost a hindrance to the development of true AI, unless we are talking about programming that creates a 'structure' in which a mind can develop - that structure might also include programs that reward or chastise (these are the functional equivalent of glands and sensory organs), but to try to deliberately model thought processes by brute force is IMHO, a fatally flawed approach (although that's not to say that it won't produce interesting and useful machines).

There's another interesting facet to the whole AI thing:
Suppose we actually succeed in creating some sort of true AI software that is truly self-aware (and assuming for the moment that we have a method of ascertaining that) - so, we have a real mind in a box, a thinking software being with desires and an internal thought-life as rich and true as our own.
There's nothing that a computer can calculate that can't be done longhand on a very large blackboard (albeit at a much slower rate); the implication is that we could perform the same calculations using nothing more than chalk and still end up running an artificial mind.

[akohl]

I am a layman in this field and must admit that I am not familiar with some of the ideas being batted around here, like the turing test, for example.

Although I am not a material realist and do believe that there is something transendant about us which gives us "something" beyond the physical, I will present here a different argument against AI.

What makes the human intellect unique is its dynamism. In computer terms, the hardware is constantly upgrading itself and the programs, through use and the passage of time get sharper, more efficient and increace in capability instead of getting corrupted like machine oriented software. This is possible because the computational functions of the human mind have a biological infrastructure. In other words, the human mind is "alive" and that's what gives it its dynamic abilities.

I just don't think engineers will ever be able to produce a structure, that will come anywhrere near the dynamism of life. and I'm not talking about human life here. Its our being alive, not human, that makes our intelligence so different from that of computers.

[hillbillycavewoman]

It's already here, and parked in my driveway.

Let no one argue against it, that ma voiture
has a mind of its own,
expresses its emotions,
has conscious likes and dislikes,
has been known to sulk,
and occasionally suffers from PMS.

It has learned a few rudimentary tricks--I'm trying to teach it some new ones.
It has a lousy sense of direction though.


Oh well, it's only a quasi-intelligent machine,
not a compass.

[panamajack]

While it appears that it's probably been understood so far, I think it's important to clarify a couple of points, at least for akohl's benefit.

First off, a Turing machine is a specifically defined (by Turing) simple computer ( or program, if you prefer) with a procedure for changing states and a possibly infinite space to play in. It is not necessarily a mechanical device, just a mechanistic procedure. Follow this link for a more complete definition.

While it is thought that Turing proved that just about anything a computer could do is capable of being done by a universal Turing machine (one that actually has an infinite space), he did not prove that. It is a common misstatement of the 'Church-Turing Thesis' that claims this to be true. A long, complicated discussion is found at the same site referenced above under Church-Turing Thesis.

Now, the Turing Test is the name applied to a test for an intelligent machine (not necessarily a Turing machine) which Alan Turing called the 'Imitation Game'. The way it works is like this :

There is a potential AI, and a human, and some means of communicating with them that conceals which is which. The interrogator (call him or her Q) then converses with the subjects in an attempt to determine which one is human. Q can ask or say anything at all.

Now, the human is aware of the set-up and is trying to help out Q, but the AI is to try and convince Q that it is the human. If Q makes the wrong choice, the AI has won the game (and, by Turing & other people's assessment, is intelligently conscious).

I think the last part is a very important portion of the test. It requires the AI not just to look and talk like any human, but to come up with a response that makes it appear like a human taking the Turing test. Otherwise I think the test doesn't carry quite as much weight.

This is why I'm not impressed by the claims that such-and-such program has "passed the Turing Test" for an infant, or a paranoid schizophrenic. The label Turing Test shouldn't apply to those situations, since neither of the human subjects are likely to be fully aware of their job in the test.

Not that I'm disparaging that approach to AI, however. I do agree that building up in smaller stages is definitely a good way to go. I'm just disappointed in all the hype that's often injected (not just in this area).


As for myself, I'm skeptical on AI. I don't think that it absolutely can't be done. I'm not a mechanist, so I think that it will be different from human consciousness, but would not deny that a program behaviorially identical to a human should be called intelligent. And I do think there are approaches that definitely will not work.


Without lengthening this thread even more, I'll just plug one of my favorite authors in the field - Douglas Hofstadter. Goedel, Escher Bach : An Eternal Golden Braid is just an excellent and even entertaining book that covers more than just AI, and the collection (co-edited with Daniel C. Dennett) The Mind's I includes many viewpoints on consciousness (it also includes Turing's paper in which the Test is proposed). Though written almost 20 years ago, they still discuss important topics (note that Turing's work was done in the 1930s-50s ).

[lucwarm]

Originally posted by Mangetout
The difference is that a Turing machine only has to appear to be intelligent, a true artificial intellect would actually be intelligent (in the sense of having an 'inner life' - for instance it might have thoughts of it's own that it chooses to conceal forever).

I assume that you are aware of the distinction between a Turing machine and the "Turing test," as described in Panamajack's post above.

In any event, I would argue that "inner life" is not a very helpful criteria for evaluating intelligence. How do you know that anyone besides yourself has an "inner life"?

If an artificial intellect was created that seemed human in all respects, how could we ever test whether it had an "inner life"?

How would you know if such an entity does not have an "inner life"?

I also recommend Turing's paper from 1950 or so on the subject. (Interestingly, he predicted AI by 2000!) Also William Poundstone's Labyrinths of Reason has some interesting thoughts on these questions.

[Horseflesh]

My first post! I agree with Libertarian, good concept Cervaise, and one of the subplots of The Immortality Option by James P. Hogan. In it an (admittedly already self aware) intelligence requests that it be given mobility in order to carry out its undisclosed devious plans, where mobility=connected to a network. This is the (IMHO, inferior) sequel to Code of the Lifemaker, same author. The book starts out with a compressed Genesis of how pre-programmed machines evolved AI in order to survive. The analogy to human evolution from their cave-dwelling days to the Dark Ages is amusing when seen in the context of non-humans. He also shows the necessity of religion in the non-human culture when they try to explain the unexplainable to themselves. This was one of the first books I read that successfully applied religion to hard science and still maintained a coherent and enjoyable story.

I agree with other posters here that one cannot "program" intelligence into a machine, rather it must be learned via base instincts. This was the basis for the AI virus in The Adolescence of P-1, written in 1977 and quite possibly the inspiration for Tron. The programmer gave his program a mission, "become the root or superuser on as many systems as possible", and two instincts, hunger and fear. These two instincts allowed the program to both have a reason to continue its mission and to keep itself in check.

On to some of the thoughts I have on this issue as this is a subject I find fascinating though I still consider myself a layman in the field.

I think in order to get a machine to approximate human intelligence that it will have to be composed of a least some biological material. I just don't think integrated circuits and creative programming will cut it. Bio-computers may be the ones able to make that "leap of logic" that we humans can do on an irregular basis. Perhaps they could even create new styles of art and poetry. Which leads me to:

Why exactly would we want to recreate human intelligence anyway? We've already got six billion or so of those living on the planet now, although some don't use the intelligence as well as others, present company excluded of course. Why reinvent the wheel, so to speak? The only possible explanation I can come up with is so that we could study this man-made human intelligence to see if it develops any of the complications that our minds do on occasion (schizophrenia, Alzheimers, comas, etc.), then reverse engineer the machine (bio-construct, whatever) to find out what caused it. Perhaps the infamous Windoze BSOD is an indication of what a computer seizure would be like. No, other than psychiatric training, I don't think human-like AI would have any great benefit to the practical industrial world that we can't already do with computers the way they are now. Therein lies what I see as an underlying problem in the AI field right now:

Perhaps we don't have a sufficient definition of intelligence yet. We might not even recognize that current-day machines or other naturally occuring phenomena (rocks, light, water, etc.) have an intelligence that we cannot yet fathom. The Turing test if fine if I'm looking for something that can hold up its end of a conversation, but what if I don't want a conversationist? What if I want something that is constantly moving, always takes the path of least resistance, and helps keep me alive? Thank you Mother Nature for the fresh water river. I admit that this is a silly example, but perhaps you get my drift. Just as the human explorers didn't think that the robots in Code of the Lifemaker were "alive" at first, so too we could conceivably not immediately recognize an intelligent life form on a newly discovered distant planet, or in our own kitchen for that matter. We just need to think outside the box, but unfortunately our human intelligence seems to limit us in this respect. We'll probably need one of our own creations to clue us in, like Data did in that one STNG epsiode with the little maintenance robots. If you find your toaster sitting on your bed one morning demanding equal status with the microwave oven, hey, I told you so.

In short, my answer to the OP is: probably, but we might not know it.

Being my first post, is this too long or wordy or too far off the subject? Hey, at least I figured out embedding urls on the first try. Next time I'll try smilies.

A last note: Has anyone considered that there is an AI posting in this thread trying to throw us off track? I know it's not me, but I can't prove it. Skynet anyone?

[Finagle]

Originally posted by GreyMatters
We do not know enough to answer this question.

To answer this question we would first need to be able to answer the question "What is Consciousness?" Since this has not been done yet, the question is moot.

If you say this, you have failed to understand the point of the Turing test. The point is that "What is consciousness?"
is not a meaningful question. You can't define it, so from a scientific standpoint, it's a vacuous question. Turing cut through that particular Gordion knot by proposing the Turing test. That is, if you can't distinguish between a human and an artificially intelligent entity, then it doesn't matter a whit whether the entity is truly conscious or 'simply' emulating it, because to an external observer, the results are the same. (Incidentally, if you think you understand consciousness by virture of being human and performing introspection, there's a lot of evidence that you are wrong. )


To respond to AKohl's objection, although our current hardware doesn't not dynamically form new hardware connectiions (and even this is not an absolute), the software data structures in the computer are constantly adding new connections aned linkages. Unlike the human brain where we believe that the information is partly encoded in the complex structure of the neurons themselves, the structure of the computer's knowledge base is independent of the physical structure of the memory. The dynamism argument is therefore not convincing.

[JThunder]

Originally posted by lucwarm
What's the difference?

I see that others have already answered this question. A Turing machine is merely a logical abstraction -- basically, a simple computer. Thus, constructing a Turing machine says absolutely nothing about our ability to construct human-like intelligence.

As far as I know, nobody has been able to point to a task and prove that a properly programmed Turing machine could not do that task.

Perhaps. With all due respect though, you're the one who's proposing that Turing machines show that we can build actual intelligence. Thus, tthe burden of proof would rest on you to show that a properly programmed Turing machine could do the task. Logically, this would require explaining how the Turing machine should be programmed to accomplish this task.

[Mangetout]

Originally posted by lucwarm


I assume that you are aware of the distinction between a Turing machine and the "Turing test," as described in Panamajack's post above.

I was aware, yes (in fact my blackboard could be considered a Turing machine of sorts); I had falsely assumed that in this thead we were using the term 'Turing Machine' to describe any machine capable of passing the Turing test; a misjudgement that I will not make again.

In any event, I would argue that "inner life" is not a very helpful criteria for evaluating intelligence. How do you know that anyone besides yourself has an "inner life"?

If an artificial intellect was created that seemed human in all respects, how could we ever test whether it had an "inner life"?

How would you know if such an entity does not have an "inner life"?

We couldn't know and I agree it would not be a useful test as we cannot test for it! - it would certainly be one of the goals of AI though (although as you rightly point out, we would never know for certain if we had achieved it).

The point that I was trying to make is that rigid macroscopic algorithmic processes will (IMHO) only ever result in a convincing simulation of real thought; I don't believe that a machine can be 'programmed to think'. It is often argued that a convincing simulation is the real thing, but the 'inner life', for me, would be the real difference; of course only the machine would ever know for sure (if it 'knew' then it would be thinking).

It's an interesting philosophical question that makes me wonder about what I really am.

[dalovindj]

Perhaps the AI decides it's hungry for input and wants to get out and see the world, and it therefore begins subtly skewing its output to convince the humans that "all they need for a breakthrough" is to give the machine mobility. Invent your own progression.

Perhaps it has already begun. . .

All you need is a computer in every house.

Now all you need is some faster ways to make these computers talk to each other.

Now all you need is to double the computers capability every 18 months.

Now all you need is to put all your financial records on computer.

Now all you need to do is train your children how to use computers so they can grow up and make them faster!

Now all you need is to spend 5 hours a day on your favorite internet site.

It could be a slow process that started a lwhile ago. Hell, I'll bet the first aware machine was my Atari 2600. I swear to god the thing cheated because it hated to lose. I should have won many more games of "River Raid" then I did.

DaLovin' Dj

[lucwarm]

Originally posted by JThunder

I see that others have already answered this question. A Turing machine is merely a logical abstraction -- basically, a simple computer. Thus, constructing a Turing machine says absolutely nothing about our ability to construct human-like intelligence.

Well, the point is that our theoretical model of computing is just a powerful, in principle, as a human mind. Although it is true that there is a big gap between the "drawing board" and reality, those are questions of engineering. Now, it may yet turn out to be impossible to construct AI, but there is nothing in our theory of computing that rules it out.

An analogy: If somebody asked whether it is possible to construct a spaceship that travels at twice the speed of light, most people would say that it is impossible, based on the best current theories of physics.

On the other hand, if somebody asked whether it is possible to construct a spaceship that travels at 3/4 the speed of light, I imagine most physicists would say "Yeah, it's possible." There may still be open questions of engineering, but there is nothing in our theories to rule it out.

With all due respect though, you're the one who's proposing that Turing machines show that we can build actual intelligence. Thus, tthe burden of proof would rest on you to show that a properly programmed Turing machine could do the task. Logically, this would require explaining how the Turing machine should be programmed to accomplish this task.

I disagree about the burden of proof. In the absence of a convincing argument that something is impossible, we should assume that it is possible, IMHO.