My post intends to be a response to a previous UD article by scordova. Scordova, who asks “should ID include AI as a form of intelligence?” and answers “I think so”, is aware to have put on the table a critical topic because himself writes:
I know many of my ID colleagues will disagree or will remain skeptical of adopting such a convention.
I am one of his ID colleagues who disagrees and I will explain why.
Scordova wrote:
So what is the evidence of intelligence? I would suggest the ability to construct artifacts or events with Specified Improbability (the usual term is Specified Complexity, CSI, etc. but those terms are too confusing).
This is an extremely reductive way to consider intelligence. Why consider only the construction of artifacts? Are people not constructing artifacts all stupid? There are countless evidences of higher activities by human intelligence. For example: the elaboration of logic, languages, mathematics, philosophy, sciences… To reduce intelligence only to its practical uses is pragmatism/materialism of the worse kind. A movement that names itself “intelligent design movement” cannot have a conception of intelligence so low.
Thus factories with robots, smart cruise missiles, genetic algorithms, bacteria, a collective network of ants, etc. can be considered intelligent systems. The problem is that we have no means of distinguishing real from artificial intelligence in any formal way.
“Factories with robots, smart cruise missiles, genetic algorithms” are not “intelligent systems” because what they produce is entirely due to the intelligence of their human designers. It is not the artificial system to be “intelligent”, rather its designer.
With no disrespect intended toward those with severe mental handicaps, yes such people are conscious, but there is a point a robotic automaton might be capable of generating more Specified Improbability than such an individual.
Conscious persons with mental handicaps, also if unable to produce “Specified Improbability”, are far more than robots because consciousness is always ontologically superior to any machine.
Some of us have imagined building robots that will land on a planet and tame it and build cities. They will act pretty much like human engineers and construction workers… Hence, the line between real and artificial intelligence gets blurred.
Again, robots build cities because they are programmed to do so by human engineers. In this case, the “line between real and artificial intelligence” is the clear hierarchical demarcation between “who programs and what is programmed”. “Who” are the human robotics engineers, “what” are the robots.
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.
On the contrary, I think that ID should carefully distinguish between real vs. artificial intelligence. (A general exhortation of Scholasticism was “distingue frequenter” in all fields.) One of the goals of ID theory is indeed to show that chance and necessity cannot produce information. Machines belong to “chance and necessity” because they are “necessitated” by their designer, so to speak. Therefore an IDer who denies the above ID proof self-contradicts.
We can even assume the process of natural selection is AI (where Natural Selection is an AI genetic algorithm in the wild), given it’s level of intelligence, we do not expect it to build extravagant artifacts.
To consider natural selection an “AI genetic algorithm” is to attribute it a merit that it doesn’t deserve. Not only natural selection is unable to build “extravagant artifacts”, it is unable to build the least artifact.
We can say an adding machine is intelligent, but we do not think, in and of itself it will build a space shuttle.
Actually I have on my desk an old mechanical adding machine. If you call it “intelligent” then why don’t call “intelligent” the reading lamp or the paperknife?
We rate the capability of various intelligence systems, and it is reasonable to affix limitations on them.
True, but here you contradict what you said before “we have no means of distinguishing real from artificial intelligence”. In fact, if we can rate various intelligences, we can see they form a hierarchy where at the top there is the real intelligence and at the bottom the artificial “intelligence”.
Whether the Intelligence that made the wonders of life is God, A Computer in Sky, Aliens, the Borg Collective, some mechanistic intelligence…it is irrelevant to the design inference. We might however be able to make statements about the level of capability of that intelligence.
Here again I see an inconsistence. I agree that we are able to grasp the level of capability of intelligence. But then, before a design inference on the universe as a whole, we cannot suppose that it was designed by “a Computer in Sky, Aliens, the Borg Collective, some mechanistic intelligence”. See here.
To sum up, to scordova’s question “should ID include AI as a form of intelligence?” my answer is: “no, we cannot consider an artificial system really intelligent”. Here I explained that real intelligence is direct connection to what I called “Infinite Information Source” (God). Here I explained that without such direct connection no comprehensibility of the world, also at the least degree, is possible. Here I explained that artificial systems (also those more sophisticated considered by AI) can show only false intelligence.
The direct connection to the Infinite Information Source (IIS) is the reason why the potentiality of knowledge of the real intelligence is infinite, as its source. No machine has this direct connection. As such the potentiality of real direct knowledge of a machine is zero. From the point of view of potentiality, the difference between real intelligence and its caricature – artificial “intelligence” – is like the difference between infinite and zero.
To deny the IIS and its connection to man is to consider man as an isolated finite system, whose potentialities are necessarily limited, due only to the configuration of its parts. This way real intelligence with its infinite potentiality of knowledge remains entirely unexplained. Said in theological terms: if man is not image of God, then man couldn’t have the potentiality of understanding he effectively has. If this simplistic materialist conception is supported by evolutionists/materialists no wonder. If it is supported by an IDer/creationist I am a bit bemused.