2 ms·
Well there's implicit and explicit. In a sense, a boiling pot of water has an implicit representation of the heat of the gas flame -- in that the state of the w
by mjburgess 5y ago
Well there's implicit and explicit. In a sense, a boiling pot of water has an implicit representation of the heat of the gas flame -- in that the state of the water tracks the state of the flame.
The question is whether the system is able to perform any explicit inference with those representations. Ie., is the representation explicit available, as a representation, eg., that this representation can be employed to imagine the world being a different way (the heart of solving problems using intelligence).
It's worthwhile distinguishing a certain pun on the word "intelligence". We don't call gangster's intelligent, and physicists dumb -- despite gangster's being much more adept at most "physical skills". The esteem for "intelligence" comes from its association with The Intellect (ie., cognition and imaginative simulation, the place of explicit representation). Not from adept procreation, at which the cornavirus is presumably then almost "the most intelligent lifeform".
An amoeba doesnt represent the world to itself. It's an interesting biochemical system which interacts with its environment in a way which seems as-if intelligent. Digital computers likewise seem as-if intelligent. Ie., they seem as-if they explicitly represent the world to themselves -- but they dont.
Many things seem to be doing what we do, many things which dont, can (ie., we can automate some problem solving so that it doesnt need intelligence). But this is only seeming. These systems are nothing like us at all, they are radically alien -- they're, in any sympathetic sense, dead. There is "nothing that they are doing" -- there's no "they" there.