In many sites on the net A-I researchers claim they can have a system of intelligence that matches all human intelligence within the next few decades. I beg to disagree, not from a religious POV, but from one of biological pragmatism. Below are my reasons for disagreement. You may agree or disagree with them and have better ideas than mine. welcome aboard. Please, no religious interpretations!
A computer or a thermostat has no reference to its own structure. In organisms almost every body cell contains total information about what the organism is and what it can become. In other words, machines lack holographic clues as to what they are.
Organisms rely on a history of internal development. Machines rely on external additions and revisions of their structures.
That organic intelligence can be compared with mechanical intelligence assumes that what is knowable can be extracted from a knower and placed in an other container.
Your thoughts?
I agree with what you have posited and would add a fourth corollary.
The machine is incapable of free expression of abstractions without prior impetus and design from organic intellectual. (i.e. an individual who sees a box, and suddenly abstracts planar existences within the dimensional aspects of a cube)
What about the strategy aspect of intellect which is another form non-algorithmic function of organic intelligence?
Also, the derivations of intellect due to environment of development, (i.e. higher order mammals such as those in the whale and porpoise families have discernably high intelligence, plausibly equal to our own, but different due to its inherent adaptation to a completely aqueous existence)
The important aspects of thinking might simply be the ability to draw conclusions based on a wide set of seemingly unrelated data. In fact, that might be what a genius does. Let’s say a marketplace genius, and not some guru type.
The average, or better yet, below average man takes learned sets of data, and does what he’s been taught to do. The military trains people in this way. The best soldier carries out his orders as they’re ordered, and not in some creative manner. In real life, lots of dumb people get into routines that they repeat endlessly until death.
So, as long as some kind of system of choice could be made for an artificial brain, then you could produce a certain kind of mind. It would be a simulation, and wouldn’t produce some quirky artist type.
Mastriani,
Exactly. Organic intelligence, which is varied according to species, runs the developmental route from simple awareness to adaptational proficiency. For humans, the genetic route is from drives and dispositions, through emotional incentives onto reasoning as as a choice of options. This active history cannot be programmed.
Mr. P.,
Emotional A-I systems exist. Rational A-I systems exist. But no such system exists or can exist that shows how one influences the other by being part of the same developmental process. Choice is at the end of organic intelligence development. One cannot simply dissect an end from its historic development and expect it to behave as if it had lost nothing in the process of being extracted.
You don’t need any of that, if you base the brain on IF/Then statements.
If the AI sees the face of Bob in its camera it will Then say “hello” and Then, check AM/PM. If AM/PM, Then good morning or good afternoon. Then, it will ask how Bob is, if it has not done so in the last four hours, Then is will check Bob’s response against the data base, and Then respond If Bob says XYZ.
It sounds funny but that’s exactly what all of us do.
If choice is a romantic notion we could all be doomed to bad choices or hard determinism. Neither is shown to be organic reality. Potential defies such assumptions; and potential is proved in growth and development. To reason is to choose. Also, recognition is not and cannot be a mechanical attribute. My computer may greet me with, “Hello, Ierrellus”, but it has no idea of who I am.
Mr. P.,
The average human you speak of experiences what I have said by virtue of being an organism, regardless of his ability or inabilty to articulate what that experience involves. You have ways and means of discovering who I am that are inaccessible to machines. What does "etherial’ have to do with anything said here?
Part of the problem with AI and AI developers is they assume too much.
Human intelligence starts with only instinct and awareness, all else is learned through manner of process.
AI developers program the unit with enormous databases, add algorithms, then equate that with human intelligence. AI does not have the awareness to cognitively choose to engage an environment for purposes of discovery.
Apparently you don’t understand the premise of discovery.
Discovery is predicated upon objective awareness of “a self” to cognitively set as an active agent within or against an environment. AI cannot replicate this, and it has nothing to do with storage or speed. AI cannot have active awareness of a self, therefore cognition of the spatial relationship of self and environment cannot occur, which obviates the possibility of discovery.
when do you think (if you think it’s a possibility) that human minds will be capable of being uploaded to computers safely and retaining personhood and idenity ?
and also…do you think that the body would also keep it’s ‘copy’ of the mind?..
then there would be 2 yous…that’d be freaky!!!
but… wow… mindf$%$
in my opinion i’s not possible to upload minds…and I believe it never will be…
but I think that AI CAN be created…though it may not be ‘human’ it still is a sort of intelligence.
A machine can be programmed to think just like a human on this issue.
Humans learn who and what they are from other humans. There has never been a case of a lone child figuring anything out.
Human learn what they are via what amounts to tautology.
A: You are Bobby.
B: What’s Bobby.
A. You.
B: What’s You?
A: No, you say me, or I. Say, I am a boy named Bobby.
B: What’s a boy.
A: You are a boy.
B: I’m a boy named Bobby.
How could you upload what we can’t even see happening?
What you’d have to do is follow someone around and get their every thought that you possiblly could, then add that to a program that was able to learn and make choices based on the person’s value system. That would be a simulation of the person.
