Should ID include AI as a form of Intelligence? I think so

Mung

Mung writes his off topic:

Satan is the great deceiver because he is the greatest concealer, the mightiest perverter of truth, the ultimate misleader, and the most convincing fraud and liar. http://www.orthodoxresearchins....._satan.htm Repent, before it’s too late.

You've got a few screws loose. That is way off topic. You're banned from these discussions and you've been told repeatedly. scordova

A complex design can be created faster by intelligent being than by random process, so the variable 'time' creeps in and when considering ET, 'time' is not invariant Let's take an example, consider that ET designs and transports a complex system at 3/5 speed of light, so the relativistic time will be calculated as: SquareRoot[1-(v^2/c^2)]. 'c' is speed of light and v = 3/5c, so we get the result 4/5. Interpretation: If it takes 5000 years by Earth time, the ET time would be 4000 years only. Of course there is the problem of calculating if there is acceleration when the complex design/process is delivered from one place to another. The relativistic time would become absurdly different. Unless you know whether a design is by ET or whether the design was delivered to Earth or wherever, there is no way to gauge and measure the complexity of a process or system. selvaRajan

Satan is the great deceiver because he is the greatest concealer, the mightiest perverter of truth, the ultimate misleader, and the most convincing fraud and liar. http://www.orthodoxresearchinstitute.org/articles/misc/evangelinidis_satan.htm Repent, before it's too late. Mung

banned Mung

Sal

For this reason, I consider the consciousness issue an irrelevancy that maybe shouldn’t be the core of ID. Central issues:

I am inclined to agree, though I am still thinking about it. I may become a serious issue some day if people start saying that computers have become conscious.

I accept I am conscious. I cannot formally prove it to others.

Right, you cannot prove it to others, but you can know it. A self-evident truth, by definition, is one that can be known but cannot be proven. It is the standard by which we prove other things. Without it, nothing can be proven.

Also, I accept a fetus is conscious. I cannot prove it to others.

Right you are. Only our own consciousness is self-evident to us. On the other hand, we can prove beyond a reasonable doubt that a fetus is a human being.

Consciousness is not self-evidently true even during the life time of individuals (myself included) that we presume are conscious. Was I conscious when I was 3-month during my mother’s term in pregnancy. I don’t know. Oddly, I think I should be in the best position to know, but I don’t because when consciousness exists is not self evidently true.

Well, consciousness is only self-evident to you if you are conscious and aware. If you are not conscious, then you cannot be aware of something that you don't have. Notice, by the way, that you can know this. You don't have to take it on faith. It is self evident to you, is it not? Many other things are equally self evident. StephenB

scordova @40, I agree with your views on consciousness and its relevance to ID. However, I do believe that there are aspects of nature that cannot be created without consciousness or a spirit. They have to do with beauty. The extravagant beauty we see in nature could not have been designed by an unconscious robot unless that robot was somehow directed by a conscious being. But I digress. Consciousness is irrelevant to ID and CSI at this juncture. People say that beauty is in the eye of the beholder. I say it's in the spirit. I see no way to obtain a formal description of beauty because it is not a property of matter. Yet, we know it when we see it. Mapou

On this point, I would disagree strongly. We know we are conscious because it is self evidently true.

I accept I am conscious. I cannot formally prove it to others. Also, I accept a fetus is conscious. I can't prove it to others. At what point is it or is it not during pregnancy? Ergo it is not self evident. Even I cannot estimate at what point during my mother's pregnancy I attained consciousness. It is not self-evidently true. My mother experienced extreme pain once related to cancer treatment. She had no recollection of that time. I believe she was conscious at that time, but no one can prove it to her, nor can she prove it to herself. Consciousness is not self-evidently true even during the life time of individuals (myself included) that we presume are conscious. Was I conscious when I was 3-month during my mother's term in pregnancy. I don't know. Oddly, I think I should be in the best position to know, but I don't because when consciousness exists is not self evidently true. I have cells in my body that did more amazing things than my conscious mind can do - like construct a nervous system. Are my cells sentient beings because they outdid my conscious mind? For this reason, I consider the consciousness issue an irrelevancy that maybe shouldn't be the core of ID. Central issues: 1. plausibility of OOL 2. plausibility of Darwinism The consciousness issue can be debated separately outside of ID as we are doing now. scordova

vjtorley @33:

6. Since there cannot be an infinite regress of explanations, we are forced to conclude that while an AI system might have generated life on Earth, it must have been sent there by a Designer who is genuinely capable of using language, in the way that we can.

From my perspective, I see no reason in principle that an AI system cannot learn to use language in exactly the way humans use it. The primary requisite for language is the ability to make analogies, i.e., to understand how a symbol can stand for something and to recognize metaphors. Analogy-making can be completely automated, in my opinion. This will happen in the not too distant future. Wait for it. Mapou

What is wrong with that? This is how science progresses.

First you have to have a theory or hypothesis, maybe do some experiments that appear to support that theory. Then, yes, others can try to repeat those experiments to see if they can be duplicated. Maybe suggest a better theory with a better fit to the data. No problem. But ID needs that first testable hypothesis to get to the scientific starting gate. Finding perceived flaws in evolutionary theory won't advance ID one bit. Alan Fox

Let’s carry on with the old shtick of attempting to knock holes in evolutionary theory.

What is wrong with that? This is how science progresses. What is unnerving is that some in the atheist/materialist camp believe that science progresses by knocking holes in religion and astrology and will not suffer science to be criticized. Mapou

We know we are conscious because it is self evidently true.

How so? Alan Fox

scordova:

No one can formally prove consciousness exists, but it is believable. The deepest truths are not provable, we have to accept them as faith axioms.

On this point, I would disagree strongly. We know we are conscious because it is self evidently true. Here we have a fundamental philosophical disagreement that seems to come up from time to time. You appear to embrace fideism, which denies the power of unaided human reason to reach certitude and affirms that the fundamental act of human knowledge consists in an act of faith. If that is your position, you will always be giving unnecessary ground to your materialist adversaries (or friends, as the case may be). Each time they press you to provide a rational defense for your most fundamental principles, you will be reduced to saying, "I can't defend it, I just believe it." They will love you for it I am sure, because they will realize that you have lost the intellectual battle even before you enter the arena. The main offensive weapon of the Materialist, and his strongest tool for defending Darwinism, is the false idea that we cannot really know anything unless it can be proven through empirical investigation. Since that doctrine cannot itself be proven through empirical investigation, it refutes itself. Still, the materialist clings to it for dear life. If you grant him that one prize, the one that really counts, he will have no difficulty in smashing your ID sympathies (and also your religious faith) into little pieces with it. StephenB

So which is it? Do we need to pin down a definition of “intelligence” or don’t we? And if we can’t agree on a definition, what is it, in your opinion, that then becomes impossible?

For me, this is a hypothetical question as I think "Intelligent Design" is an empty concept. But not having a theory or a concept or any operational definitions for any of the concepts you are trying to demonstrate is a handicap if you are out to show ID is a scientific pursuit. I don't dogmatically say you can't carry on and look for well, whatever it is you are going to look for. I would admire anyone that actually got going and made an effort. As I said, though, in those circumstances I can't see where you would start. You and anyone are welcome to start, however, with all god speed from me.

BTW, you presumably are aware that evolution (in its materialistic sense) operates as a foil for purposeful design (and vice versa). Critiquing the many failings of evolutionary theory is certainly a legitimate exercise in its own right; it is additionally relevant to the extent it is — as so often is the case — wielded as an attack against the idea of design in the world. Darwin was a prime example of this, and his followers are still at it.

It is probably true to say that a large part of the criticism directed at Darwin's ideas was religiously motivated and I guess it's still true but to a much lesser extent. The biological sciences seem to be a particular bête noir for some religious groups but it seems inevitable when dogma is allowed to outweigh facts. The Dalai Lama wrote:

My confidence in venturing into science lies in my basic belief that as in science so in Buddhism, understanding the nature of reality is pursued by means of critical investigation: if scientific analysis were conclusively to demonstrate certain claims in Buddhism to be false, then we must accept the findings of science and abandon those claims.

Dalai Lama: The Universe in a Single Atom Alan Fox

Hi Sal, Thanks for a very thought-provoking post. The questions you raise are valid ones, and they do need to be confronted. A few thoughts: 1. The question of whether the Intelligent Designer is conscious is a red herring. Consciousness is commonly envisaged in phenomenal terms, as a subjective feeling of "what it is like" to be an entity of a certain kind (e.g. a bat, a dog or a human being). Many Christians would adamantly deny that God is conscious in that sense: they would say that He has no "inner life," and they would reject as anthropomorphic the very notion that He does. Thomists would fall into this category, for instance: for them, God is in no way passive, so He has no experiences of anything. His intelligence is entirely active. Other Christians (e.g. Richard Swinburne) would reject this view. I think it would be unwise for the Intelligent Design movement to take sides in this dispute. 2. For me, an important question is whether the Intelligent Designer has foresight or not. For instance, James Shapiro's non-Darwinian mechanism for evolution is "feeling-intelligent" but not "seeing-intelligent": it can select the best alternative from those that are available, but it can't "look ahead" to the long-term future. 3. Describing a colony of bacteria or an ant hill as intelligent is misleading. They lack foresight. They can exchange information very efficiently, but they can't plan ahead. 4. Another important question is whether the Intelligent Designer is capable of using language or not. Only a language-using entity can design a code, such as we find in living things. And only a language-using entity can answer questions such as: "Why did you design it that way, and not some other way?" 5. AI systems are programmed in some kind of language, but are not able to use language in the flexible way I outlined above: that is, even if they can be programmed to converse on a limited range of topics, they are incapable of critiquing their own internal logic. Hence whatever "intelligence" they may be said to possess is secondary. 6. Since there cannot be an infinite regress of explanations, we are forced to conclude that while an AI system might have generated life on Earth, it must have been sent there by a Designer who is genuinely capable of using language, in the way that we can. 7. Finally, the fact that we now know that even the multiverse must have had a beginning means that atheists cannot take refuge in an infinite regress of robots, or a "Robot in the sky." My two cents. vjtorley

Alan Fox @30: So which is it? Do we need to pin down a definition of "intelligence" or don't we? And if we can't agree on a definition, what is it, in your opinion, that then becomes impossible? ----- BTW, you presumably are aware that evolution (in its materialistic sense) operates as a foil for purposeful design (and vice versa). Critiquing the many failings of evolutionary theory is certainly a legitimate exercise in its own right; it is additionally relevant to the extent it is -- as so often is the case -- wielded as an attack against the idea of design in the world. Darwin was a prime example of this, and his followers are still at it. Eric Anderson

AI and robots that produce CSI are not natural processes. Thus they do not threaten ID. Frankly, I don't see any problem for ID. Maybe scordova and others can explain further. Mapou

Thus we get sidetracked on discussion about NFL, rather than seeing clearly that mindless OOL and Darwinian evolution in the wild will not work.

I had to smile at this bit of Sal logic. Lets not get sidetracked on whether proponents of "Intelligent Design" will ever present a theory or hypothesis, or even a working definition of "intelligence". Let's carry on with the old shtick of attempting to knock holes in evolutionary theory. Yeah, that'll work! :) Alan Fox

Sal @23:

It raises, at least in principle, the possibility, non-conscious entities making CSI. Why do I bring this up, you can expect enemies of ID to throwing stuff like this in our face in the future. I’m surprised they haven’t said as much until now. But I’m putting it on the table before they say, “We told you so.”

This is nothing new (your particular effort to resolve it may be, I don't know). Materialists have been claiming for a long time that non-conscious entities can make CSI. Ever since Dawkin's discredited Weasel "program," Avida, and all the others. The claim is always the same: We can get CSI from purely natural processes, through Darwinian trial and error. This is the heart of Elizabeth Liddle's concern about ID, for example. Again, I think you need to be more careful to distinguish between a surrogate reproducing CSI and the original source of the CSI. It seems that is the key to the discussion. Eric Anderson

Some will ask, "well if not God, then who created the first life". Some have seriously said it is the great computer in the sky -- Seth Lloyd, for example. Fred Hoyle (intelligent universe), and Frank Tipler (sort of, non-traditional conception of the Christian God). The multiverse is sort of a mindless OOL mechanism. scordova

Isn’t it the case that ID is about the source or cause of the creative conception that informs CSI rather than the method by which it is produced? If so, isn’t it the case that a robot could never be the source of that creative conception? Isn’t ID less about playing the music and more about writing the score?

There are several issues within ID: 1. mindless OOL 2. Darwinian evolution 3. the necessity of conscious intelligence to create life #1 and #2 can be dealt with without having to answer #3 first. Refute #1 and #2, and #3 can be accepted by many with a small step of faith. I don't know whether #3 can be formally demonstrated, but I accept it as true on reasonable grounds. No one can formally prove consciousness exists, but it is believable. The deepest truths are not provable, we have to accept them as faith axioms. I do believe the ultimate source of all CSI comes from God, but we keep hearing ID proponents say, "CSI cannot increase except via conscious intelligence", to which I say,

"maybe on average that is true, but if that is the case on average, that means there are some processes that are above average, like chess playing machines! Average means there are things above and below average, and hence, the force of NFL claims can always be objected to on the grounds we are seeing some deviation from the mean."

Thus we get sidetracked on discussion about NFL, rather than seeing clearly that mindless OOL and Darwinian evolution in the wild will not work. The mission of UD is serving the ID community. If ID sympathizers like myself and Mapou raise these issues, how much more will the opponents of ID? scordova

scordova:

It raises, at least in principle, the possibility, non-conscious entities making CSI. Why do I bring this up, you can expect enemies of ID to throwing stuff like this in our face in the future. I’m surprised they haven’t said as much until now. But I’m putting it on the table before they say, “We told you so.”

Isn't it the case that ID is about the source or cause of the creative conception that informs CSI rather than the method by which it is produced? If so, isn't it the case that a robot could never be the source of that creative conception? Isn't ID less about playing the music and more about writing the score? StephenB

BTW, are you claiming that we cannot look into effects of intelligence, or consider whether intelligence is different from purely natural processes unless we can come up with a comprehensive, mutually-agreed-by-all definition of “intelligence?”

Nope! Go ahead. Who's we in this instance? Alan Fox

fossil @14:

That being said, Mapou, I am a little uneasy with what you believe is intelligence. If intelligence is a matter of sensory perception working through some logic circuitry then the computer on my desk is intelligent. I then can ask does a creature such as a spider display intelligence when it builds a web. From what I know the spider is completely preprogrammed and can do the job the minute it is born. If a boulder rolls down a hill and is deflected when it hits an object is that intelligence? After all it could be considered a form of preprogramming like crystal growth. But I don’t think such things are true intelligence.

I see your point but, IMO, whether or not the behavior is preprogrammed, all of those things are intelligent. It's a matter of degree of sophistication. What we're after here is what some refer as "true intelligence". Humans have it and many animals have it too. It involves the ability to learn patterns and sequences of pattern through sensory experience. It's all in the timing of events. This kind of learning is self-programming. From this ability comes the making of predictions, anticipation, planning and goal-oriented behavior. The latter is the holy grail of robotics, so to speak. All of those things can be done by a computer program.

The term AI (artificial intelligence) was coined when people in computer science began to mimic the human mind. The outcome of that were expert systems which only work in a very limited capacity. It is true that to the average person such a system has the appearance of intelligence but the rules are generally inflexible. Even when the program is able to alter itself there are usually set boundaries beyond which it cannot go unless one wishes to dabble in pure statistical probability.

Yeah, the early AI pioneers, following in the footsteps of Alan Turing, believed that intelligence was just the manipulation of symbols. Based on that assumption, they could produce nothing but limited domain AI programs, the expert systems that you mentioned. It took nearly half a century to put all that nonsense to rest. Nowadays, the buzzwords are "machine learning", "statistics" and "prediction". This is all well and good but the problem is that most of the AI community has embarked on yet another wild goose chase by adopting Bayesian statistics as the basis of perceptual learning. The Bayesian Brain hypothesis is turning out to be pretty successful and is the basis of impressive progress in such areas of speech and visual recognition. However, IMO, the Bayesian approach is wrong. There are aspects of perceptual learning that it cannot handle. One such problem is the cocktail party problem, i.e., the brain's ability to focus on one speaker in a roomful of other speakers and extraneous noise.

Compare that to the human mind that can appreciate unquantifiable data – that can create things without a precedent. How do you program that? Perhaps what I am saying is that to me real intelligence is not mechanical, it goes way beyond that.

Well, I don't think so. We can write programs that can deal with uncertainty and probabilistic phenomena in the sensory space. The assumption that we can create things without precedent is not really true. Everything new is based on older things. It's all part of the learning process. In the end, it's all about the ability to recognize and reason about causes and their effects. Mapou

Creating CSI will be a routine capability of our intelligent machines. In a way, we can already see CSI in existing intelligent programs such as a computer chess program. A chess program can anticipate a complex series of future moves and counter moves and act accordingly.

If we build robotic planetary probes capable of doing construction with planetary materials, this will be true. Because there is substantial delay in transmitting information between places like Earth and Mars, these robots will need to have autonomy similar to humans. It raises, at least in principle, the possibility, non-conscious entities making CSI. Why do I bring this up, you can expect enemies of ID to throwing stuff like this in our face in the future. I'm surprised they haven't said as much until now. But I'm putting it on the table before they say, "We told you so." With respect to chess, the computers don't store all possible moves, they make it up on the fly with heuristics. I do not know off hand whether the heuristics in play now cover all possible moves. In the opening of his NFL book, Dembski said intelligent agents can work through surrogates. Indeed this is true, but the problem is the surrogate act with such efficiency, for certain tasks they exceed the designer in making CSI! Chess programs are a good example. scordova

Sal, I very much liked your many topics in the past (I sympathize with the YEC thinking) but I am going to be kind of critical of both the wording of your topic and some of your statements

Thank you for the kind words and also the criticism. It's a topic I think the ID community needs to discuss. I'm putting it on the table. I have my opinion, and I think others need to weigh in. Thanks for your thoughts. Sal scordova

Alan Fox @16:

Until someone can think of a working definition for intelligence that makes some sense, I can’t see how you can even start.

I think the AI community, especially people like Jeff Hawkins (he's an atheist/materialist but he's pretty knowledgeable about AI), have a pretty good handle on what intelligence is. Intelligence is primarily the ability to learn patterns and sequences of patterns and to predict the outcome of various sequences of patterns based on experience. Prediction is the key because it allows the intelligent agent the power to anticipate. Anticipation is the essential ingredient needed to create CSI. It is the sine qua non of motor learning and adaptation (appetitive and aversive behavior). Creating CSI will be a routine capability of our intelligent machines. In a way, we can already see CSI in existing intelligent programs such as a computer chess program. A chess program can anticipate a complex series of future moves and counter moves and act accordingly. One of the biggest problems with Darwinian evolution is its inability to anticipate the future. No animal or human could survive in their environment without this ability. Mapou

Alan Fox: BTW, are you claiming that we cannot look into effects of intelligence, or consider whether intelligence is different from purely natural processes unless we can come up with a comprehensive, mutually-agreed-by-all definition of "intelligence?" I hope you aren't saying this. But if so, please do tell all those poor folks working in areas of psychology, counseling, artificial intelligence, criminal investigation, etc., all of whom regularly use the concept, and none of whom have a be-all-and-end-all definition. The general everyday dictionary definition is quite adequate for most purposes. The definitional game -- as important as clear definitions are -- is so often used just as an escape valve to avoid dealing with the real issues. Eric Anderson

Alan Fox @16: Yes, we've looked at this antenna before. No-one doubts that trial-and-error approaches can be useful for looking at simple problems, particularly where the constraints and parameters -- as in this case -- were carefully input into the original conditions and the outputs were carefully selected in a steady march toward a very specific goal. The whole process could have been done by an engineer in a lab; computers are just helpful to speed up the process in this kind of simple case. Incidentally, the Wikipedia claim that the process mimics Darwinian evolution (which by definition has no constraints, parameters, goals or desired outcomes), is silly propaganda. Eric Anderson

Incidentally, it seems to me on a quick read that part of your issue is that you define “intelligent” in reference to the ability to create CSI, which unfortunately ends up being somewhat circular.

Somewhat circular? :) Alan Fox

I see fossil has made a similar point. Apologies to fossil for missing it on an over-brief scanning of comments. Alan Fox

Sal This antenna would appear to fit your requirement for "specified probability". In case you want to ask me again, BTW, the human designers and builders of the space shuttle could described as comparatively intelligent WRT the general population of the USA, as I suspect many of them would perform comparatively well on intelligence tests. All that really elucidates is how well a particular candidate performs at that intelligence test. Of course the elephant in the room is whether intelligence or cognitive ability can reside outside a human (or other sentient animal or being) brain or outside a computer modelled on a human (or less complex) brain, the construction of which I see no fundamental physical barrier which would prevent that achievement some day. Does intelligence exist independently? Can supernatural intelligence be inferred or extrapolated from the known examples of real sentient beings? Until someone can think of a working definition for intelligence that makes some sense, I can't see how you can even start. Alan Fox

Sal, I very much liked your many topics in the past (I sympathize with the YEC thinking) but I am going to be kind of critical of both the wording of your topic and some of your statements. I don’t like the title you choose – but sure you were free to choose it. I suggest that maybe a proper question to ask might be: Q1. Can Intelligent Design be produced by Artificial Intelligence Agents? And before even bothering to answer question Q1 above we need to ask some preliminary questions: Q2. Do real Artificial Intelligence Agents exist? (or will they exists in the future at least?) Even if we assume that the answer to question Q2 above is “YES”, here is the next question: Q3. Can (presumptive) Artificial Intelligence Agents be created by anything else but by Real Intelligence Agents? I would say that the three questions above may indicate that the question posed by your title is somewhat superfluous or at least it may get a more precise and beneficial expression. Overall I think asking a question as that in the topic doesn’t help in any way the “search for truth and clarity” in the Intelligence Design realm. Not because this will create who knows what new difficulties for ID but just takes us on a tangent, into the fuzzy realm of defining intelligence, etc – as suggested by Eric Anderson at #1. Fundamentally materialists and evolutionists base their religious belief system on the following proposition (“credo”): Material nature (matter, energy, laws of physics and chemistry and happenstance) creates Artificial Intelligence Agents that, in turn, create (and expand) Nature.

Thus factories with robots, smart cruise missiles, genetic algorithms, bacteria, a collective network of ants, etc. can be considered intelligent systems

I think we mix here different categories. I see in this list: # factories with robots, smart cruise missiles, genetic algorithms = these are artifacts produced by intelligent agents (humans) and complex specified systems – but not necessary “intelligent systems" – i.e. they are NOT Intelligent Agents” # bacteria – this is an artifact produced by an unknown (for some) Intelligent Agent – and exhibit a very high level of complexity – and maybe can be classified as an “intelligent system” but not as an Intelligent Agent. Keeping my promise to be critical: I don’t like your term of “Specified Improbability”. I don’t think it helps ID understanding and it is not better then Specified Complexity, CSI, etc. As a software engineer with some exposure to artificial intelligence, I am rather skeptical of any real present or future Artificial Intelligence Agents. It seems to me that the business of Intelligent Agents is the exclusive domain of God. And we humans are real Intelligent Agents just because He chose to create us in His image. When I talk of a real Artificial Intelligent Agent I am not thinking of an expert chess player or an expert system in cardiac illness diagnostics, but rather a multi-lateral intelligent agent that can mimic the human intelligence. InVivoVeritas

If all of you don’t mind having an old relatively uneducated brain dead old man put his two cents worth in I would like to suggest that sooner or later the ID community is going to have to define what they mean by intelligence. If we allow atheistic materialism to do it they will no doubt define the term so loosely that anything can be intelligent just like they are attempting to do with the definition of life. That being said, Mapou, I am a little uneasy with what you believe is intelligence. If intelligence is a matter of sensory perception working through some logic circuitry then the computer on my desk is intelligent. I then can ask does a creature such as a spider display intelligence when it builds a web. From what I know the spider is completely preprogrammed and can do the job the minute it is born. If a boulder rolls down a hill and is deflected when it hits an object is that intelligence? After all it could be considered a form of preprogramming like crystal growth. But I don’t think such things are true intelligence. The term AI (artificial intelligence) was coined when people in computer science began to mimic the human mind. The outcome of that were expert systems which only work in a very limited capacity. It is true that to the average person such a system has the appearance of intelligence but the rules are generally inflexible. Even when the program is able to alter itself there are usually set boundaries beyond which it cannot go unless one wishes to dabble in pure statistical probability. Compare that to the human mind that can appreciate unquantifiable data - that can create things without a precedent. How do you program that? Perhaps what I am saying is that to me real intelligence is not mechanical, it goes way beyond that. I have said in other venues that the difference between an ape and a man is that humans can look at a sunset and appreciate its beauty while the ape sees it and starts looking for a place to sleep the night. While it can be argued that the ape has intelligence, for the purposes of ID I prefer to define intelligence beyond the mechanistic thinking of apes. Any ape can hit the keys on a typewriter but it takes a human to make a meaningful sentence on it, to create specified information. That I believe is the boundary of real intelligence. Therefore if intelligence is merely mechanical operation then evolution has a point in saying that it has intelligence and chooses to evolve into higher forms and we would have basically lost before we begin. fossil

Everyone can tell the difference between a living being, people, critters, and a machine or plant. One group has feeling and one doesn't. Now if feelings is not good enough then the better word is thoughts. M<achines have no thoughts but only act upon memory. Plants are mere memory operations. It all comes down to the anatomy of the soul. The gOd idea of living beings. So it might be too complicated for people to figure out its parts but machines doing actions from programs would be like saying a rock falling off a cliff is thinking all the way because its operating gravity and breakup principals at the bottom. Its about thoughtfulness as opposed to intelligence. Robert Byers

Sal @6: Thanks. Some good food for thought that I'll have to think about for a while. Just a quick comment before retiring for the night: I don't view the regression complaint as particularly compelling. Regardless of whether we are seeing the effect of the original designer or a surrogate, it is pretty clear that an original was required. We don't have to think of anything as complicated as a cell. I can write a program to go grab a text file of War and Peace and reproduce it. We could then say -- through some hyper-technical definitions -- that my program "created" CSI and is therefore "intelligent," but that stretches the meaning of both the word "created" and the word "intelligent." Similarly, the CSI that is "created" by a cell, presumably already existed in the cell. Do we really want to go down the path of arguing that reproducing something is the same as creating it? Additionally, your approach seems to be require that we determine, as an initial matter, that natural processes, such as, say, natural selection, are not up to the task. Perhaps we cannot completely get away from that in practice -- after all, the explanatory filter defers to natural explanations, should they be up to the task. So perhaps the negative side of the ID argument is largely the same under your proposal.* But I'm wondering if your approach throws out the positive side of the ID argument? Namely, we know from routine experience that intelligent agents create CSI. Incidentally, it seems to me on a quick read that part of your issue is that you define "intelligent" in reference to the ability to create CSI, which unfortunately ends up being somewhat circular. Maybe you are right, but I've never felt it was necessary to define intelligence solely with respect to its ability to generate CSI. Maybe the other aspect of intelligence is the conscious/willful part of the discussion you are trying to avoid? I guess I'm not too bothered by that aspect, because in my experience the existence of conscious, will-driven, intelligent beings is so obvious that the only people who battle against it are those who are intent on avoiding rational debate. Anyway, I'll chew on it a bit more. Interesting stuff. ----- * BTW, I'm wondering if your approach adds another complication to the argument. Specifically, if we have to ask, in each case "can x turn into y over time," that would seem to put the onus back on proving that it can't happen in each individual situation. Worse, it might be the case that x can turn into y over time because it was pre-programmed or front loaded to do so. We might then come up with the wrong answer unless we are careful. Ultimately, I'm not sure how you get away from referencing the original source of the CSI, not just the proximate instantiation or reproducer of the CSI. Eric Anderson

scordova @6:

The ID arguments are getting bogged down in discussion of consciousness when the real issue is the capacity of simple undirected processes to generate the extravagance in biology.

I don't think robots can create the extravagance we see in nature on their own. We observe amazing beauty in nature. Robots will not nor can they have a sense of beauty unless they are trained to recognize a limited number of beautiful things by a conscious intelligent being. Beauty is a spiritual phenomenon. It is not a property of matter. Mapou

selvaRajan @9, I don't know for certain. I only know that particle interactions at the quantum level are probabilistic. I speculate that the spirit can sense and affect the probabilities and, in so doing, change the outcome of certain interactions. Mapou

Hi Mapou @8,

From my research, I have come to believe that the spirit that inhabits the brain can interact with it via certain atomic configurations at the quantum level

Could you elaborate on that? What constitutes the Spirits? How would the constitution of Spirit interact with Quarks? selvaRajan

Eric Anderson:

You have some interesting thoughts. One question: Can intelligence arise due to a particular arrangement of matter? (Either by careful planning and design, or by evolutionary trial-and-error?)

Intelligence must be either designed or created. There is absolutely no way it can happen by chance. One may ask, who designed or created God? To that I would reply that God created his own brain and body. I believe that certain entities in the spiritual realm can create matter, including their own intelligence. Who created spirits? Nobody. They just are. The spiritual realm can neither be created nor destroyed. Only in the physical realm can things be created or destroyed. That's my take on it.

The reason I ask is that the essence of the evolutionary story is that particles bumped into each other for a long time and eventually, from that chaotic nonsense, emerged sense; from the unintelligible, intelligence; from meaninglessness, meaning; from unconsciousness, consciousness. I’m wondering whether all of this can in fact arise through a particular arrangement of matter, or whether at least some of it lies beyond matter itself.

I don't believe for an instant that consciousness arises from matter. That's absurd. It takes two things to have consciousness, a knower and a known. The two are opposite and complementary by definition. That is to say, the knower cannot be known and the known cannot know. The known is physical and the knower is spiritual. Not even God can know a spirit directly. This is why the scriptures teach us that spirits must tested because they are known only by their actions, i.e., indirectly. The idea put out by the likes of Douglas Hofstadter and many others in the materialist camp according to which consciousness arises from some kind of strange recursive loop is ridiculous to the extreme. Those guys know not what they're talking about.

You seem to be distinguishing between consciousness, for example, and intelligence. I guess if we defined intelligence as broadly as “being able to select from a range of possibilities,” then lots of things might qualify as intelligent. I tend to view intelligence in a more narrow light, including the ability to choose whether or not to follow the built in decision tree, the ability to deviate from pre-programed solutions, the ability to deal with new, novel situations, the ability to know that a decision is being made (though perhaps this last one slips close to the idea of consciousness). In any event, if we take the view that intelligence and/or consciousness can arise through a particular arrangement of matter, then — at least in that limited sense — our “explanation” for intelligence/consciousness would not be different from that of the materialist evolutionist. We might dispute whether natural processes can indeed create such an arrangement of matter, but on the question of whether matter can give rise to meaning, intelligence, consciousness, etc., the viewpoint would be the same.

The materialists don't have a clue. They are lost in the wilderness. True AI will never come from the materialist camp because they have willingly blinded themselves. In my opinion, only the spirit that inhabits the brain can choose to deviate from its pre-programmed motivation. It is the spirit that chooses and sets likes and dislikes. The spirit can, for example, choose to focus on music and the arts. The brain is not pre-programmed for liking art and beauty. Robots will never develop an affinity for the arts other than what they are conditioned (trained) by us to focus on. And they will do it only to get a reward from us. The appreciation of beauty is not a material thing. It's a spiritual thing. The oft repeated notion that intelligent machines will rebel against humans and take over the world is a materialist's stupid wet dream. They understand absolutely nothing about motivation or where it comes from. Humans will design intelligent machines to harm other humans and/or other machines. The enemy is not the machines. It is us. We are the evil ones. The machines will always be an obedient tool that can be used for either good or evil, regardless of how intelligent they are. It's scary though. Those who don't have intelligent robots will be at the mercy of those who do. From my research, I have come to believe that the spirit that inhabits the brain can interact with it via certain atomic configurations at the quantum level. But this is true only in certain parts of the brain. The cerebellum, for instance, a huge sub-area of the brain that handles routine tasks such as posture, walking and balance, is never conscious. It is a complete automaton. Mapou

I was simply offering alternative means of arguing for design, and trying to find a line of argument that is less vulnerable to some of the pitfalls of invoking conscious intelligence or the pitfall

What you would be doing is parceling the complexities of process into another complexity (ET). You can sit back and argue that my job is done. Let some one explain origin of ET. Darwinists have parceled off the origin of life to abiogenesis but the point remains - we still have to account for inanimate origin of life. selvaRajan

I’m not completely sure where you are heading with your comments

Cells make other cells, a cell contains CSI, therefore a cell makes CSI, therefore a cell is "intelligent" by ID definitions. There maybe regress issues in play, but we can't deny if a cell contains CSI, and if its parent cell created it, it is hard to say the parent cell doesn't possess intelligence proximally at some level. Do we always have to invoke regress arguments to some ultimate cause? We can if we wish, but we don't have to in order to pose other interesting questions like: "can this single-celled creature evolve into a bird?" If not, then we don't even need to invoke regress arguments, we know a greater intelligence than natural selection in the wild was necessary. We bypass the need to go to regress arguments (like the cell had to be designed). Instead we could argue directly, the bird had to be designed by something other than Natural Selection. The ID arguments are getting bogged down in discussion of consciousness when the real issue is the capacity of simple undirected processes to generate the extravagance in biology. I think genetic algorithms can generate Specified Improbability (or CSI), and again we get bogged down in discussion of whether it is possible or not. Dogs are presumably intelligent creatures. I don't expect them to have the ability to build space probes. Hence by supposing for the sake of argument a system is "intelligent" we can ask "is it intelligent enough, and does it have the means to use its intelligence?" Even granting a genetic algorithm or natural selection in the wild has some level of intelligence, as we see it in operation today, it will not evolve monarch butterflies from bacteria. Plenty of lab evidence to the contrary. Bacteria can be said to have some level of mechanical intelligence, for that matter every cell. They do make specified complexity, but not the sort that will create the extravagance in biology. NS works in the wild, but it doesn't have the sort of fitness functions that will enable it to evolve extravagance. Even supposing NS is a primitive mechanical intelligence, it is not intelligent enough. I was simply offering alternative means of arguing for design, and trying to find a line of argument that is less vulnerable to some of the pitfalls of invoking conscious intelligence or the pitfall of ignoring the Specified Complexity (Improbability) that is generated by supposedly non-sentient systems (like cells). CSI can be generated by non-sentient entities. We get into complications when we insist otherwise. In Dembski's book, No Free Lunch, he refers to these non-sentient entities as "surrogates" (in the opening of the book). Dembski makes the case these surrogates can't spontaneously emerge. I agree with that, but we don't always have to appeal to that claim. scordova

Great idea Sal, YECs can claim God created everything so no need to explain complexities, IDist can claim ET created everything and 'We live in a Sim' guys too can claim the same. All evolutionary scientists, physicists, cosmologists who have those views can pack up and go home with pink slip in their hand, and join SETI in the search for ET. Any kind of progress will be left to Darwinists. selvaRajan

Mapou: You have some interesting thoughts. One question: Can intelligence arise due to a particular arrangement of matter? (Either by careful planning and design, or by evolutionary trial-and-error?) The reason I ask is that the essence of the evolutionary story is that particles bumped into each other for a long time and eventually, from that chaotic nonsense, emerged sense; from the unintelligible, intelligence; from meaninglessness, meaning; from unconsciousness, consciousness. I'm wondering whether all of this can in fact arise through a particular arrangement of matter, or whether at least some of it lies beyond matter itself. You seem to be distinguishing between consciousness, for example, and intelligence. I guess if we defined intelligence as broadly as "being able to select from a range of possibilities," then lots of things might qualify as intelligent. I tend to view intelligence in a more narrow light, including the ability to choose whether or not to follow the built in decision tree, the ability to deviate from pre-programed solutions, the ability to deal with new, novel situations, the ability to know that a decision is being made (though perhaps this last one slips close to the idea of consciousness). In any event, if we take the view that intelligence and/or consciousness can arise through a particular arrangement of matter, then -- at least in that limited sense -- our "explanation" for intelligence/consciousness would not be different from that of the materialist evolutionist. We might dispute whether natural processes can indeed create such an arrangement of matter, but on the question of whether matter can give rise to meaning, intelligence, consciousness, etc., the viewpoint would be the same. Eric Anderson

scordova:

I will say this, it appears to me, the intelligence that makes a creature as awesome as a monarch butterfly seems far beyond the collective intelligence of humanity, and certainly far above the “intelligence” of natural selection in the wild.

There is no reason, in principle, that our collective intelligence, i.e., our knowledge, understanding and tools (including our future intelligent machines) cannot advance in say, a few 1000 years or even less, to the point of being able to design and engineer something as awesome as a monarch butterfly. We, too, are gods. As a Christian (an unorthodox one, I admit), I don't believe God is just one individual but a collection of many beings or, if you wish, a united civilization consisting of many intelligent specialists (or elohim as Genesis calls them). I can see evidence for this in the vastly different life forms and design styles that are observable on earth. Humanity, too, has a collective intelligence. No single human being could have designed the computer, the internet, the modern automobile, radio or television. Many individuals have contributed over the years. Likewise, in the future, we will have huge numbers of robot specialists that, together, form a sort of collective intelligence. Mapou

scordova, I believe that, in the not too distant future, there will be no fundamental difference between natural intelligence and artificial intelligence other than that the former is composed of living tissue and the latter is not. Intelligence is primarily the ability to learn from sensory experience and to make predictions (anticipate). From this ability, a machine will develop motor skills and use them to achieve short and long term goals based on that ability. The main difference between human and future intelligent robots is this: The former is self-motivating (i.e., humans set their own likes and dislikes) whereas robots must be pre-programed with a set of motivations. But there will be no difference in the way they go about seeking "rewards" and avoiding "punishment". In fact, given that intelligent robots will be less distracted overall than humans, they will be much more effective at achieving their goals. Given the above, there is no reason to think that we could not send a legion of intelligent robots to terraform other planets and prepare them for human habitation. I predict that, within the next 20 years or so, human labor will become obsolete and humanity will have to drastically change its economic systems. It will be neither capitalism nor socialism because both are based on human labor. Question: Will intelligent robots be conscious? Answer: Absolutely not. The problem I see is that many will swear that intelligent robots are conscious and will insist that they be given legal rights similar to humans. I cringe at the thought. Mapou

Sal: You have a number of interesting thoughts. Taking a couple in turn:

. . . the usual term is Specified Complexity, CSI, etc. but those terms are too confusing.

Nah, the terms are pretty clear, except to critics who are intent on misunderstanding them. :)

Hence, the line between real and artificial intelligence gets blurred. . . . From an empirical standpoint, I don’t think it does ID much good to try to distinguish the outcomes of real vs. artificial intelligence, since we can’t formally demonstrate one from the other anyway, at least with regard to Specified Improbability.

I think there is an important point lurking here. One of the frequent and vocal complaints against ID is the fact that no comprehensive definition of "intelligence" exists. Such a complaint is true, but largely meaningless. No-one has a comprehensive be-all-and-end-all definition of "intelligence" -- not neurology, not psychology, not AI, no-one. Yet that does not prevent those disciplines from being "science," nor does it prevent them from doing good work. I am personally skeptical that a final, definitive definition of "intelligence" is even possible. And it would behoove ID proponents to not get bogged down in definitional skirmishes when the broader claims and key points of ID can be made perfectly well by using the word in the everyday ordinary context of a basic college dictionary, particularly in light of its Latin etymology: "to choose between." So, Sal, I'm not completely sure where you are heading with your comments, but I do agree that we ought not get bogged down in whether a particular intelligence is 'artificial' or 'real' for purposes of most discussions. ----- To be sure, there is value in distinguishing between an expression of intelligence and the source of that intelligence. In the case of your adding machines -- arguably even in the case of your terraforming robots -- any "intelligence" they exhibit is pre-programmed into their systems by the "real" intelligent agent, the programmer. That, of course, is not what AI proponents have in mind with the goal of creating a truly artificial intelligence. Personally I think the jury is still out on whether it is possible to create what we would recognize as true intelligence by putting molecules into a certain configuration. At the very least, the past several decades have demonstrated that the difficulty of the endeavor has been vastly underappreciated. ----- P.S. I'm also not a fan of attributing intelligence to things like the Internet or groups of individuals. Those are merely interactive expressions of numerous individual intelligences and we abuse words and confuse discussions by referring to such things as "intelligent." Eric Anderson