Artificial Intelligence is a contradiction, intelligence is natural to the extent that all humans Just Us Florida Gators Bowl Champions shirt, and most mammals have it to some extent. Machines do not. We do not know how intelligence works, how the mind collates input from the sensory system, memories, instincts, and learned information to form thoughts and ideas. What truly sets us apart from other mammals, however, is consciousness, self-awareness, and awareness of our relationship to our environment. Computers process information by parsing vast quantities of data incredibly quickly, applying filters, and producing a conclusion based on statistical analysis. This is not intelligence. I have said many times the only way we will even be able to say machines are genuinely intelligent is if we radically redefine what we mean by intelligence. Causality is a model. And like all models, it is wrong. But like some models, it is useful. I suggest that causality is really two models, each useful in its own way.
By “wrong”, I mean incomplete. All models must be incomplete – that’s what model means. There is no way I can model all the interactions between photons and silver atoms that result in me seeing my reflection in the mirror. But if I simply say that the mirror causes the beam of light to be reflected – that works. Never mind that there is no way a silver atom could know which direction to re-emit the photon it just absorbed; the overall result is stable and reliable. The first model of causality is like the reflection of light: an event incites vast numbers of small interactions, which produce a reliable net effect. The effect reliably follows from the inciting event, so we say the effect was caused by the event. The cause can be treated as one single interaction. Many examples of this can be used to prove the usefulness of the model, without changing the unreality of the model. Both models involve a gross simplification of complex interactions. Both are useful. The first is collectively useful in allowing anyone to make predictions about the world that have a good chance of coming true.