“AlphaGo has demonstrated that a machine can learn how to do things that people spend many years of concentrated study learning, and it can rapidly learn how to do them better than any human can.”

— Semantics derived automatically from language corpora contain human-like biases; Aylin Caliskan, Joanna J. Bryson, Arvind Narayanan.

This statement is an explicit example of a catastrophic category error, for »it is formally impossible for a machine to play a game. Machines execute algorithms according to mathematical processes and using data housed in electronic forms. This is not the same category of activity as ‘play’ and never can be.

No machine can ever ‘learn how to do things that people spend many years learning’, nor can any machine ‘learn’ at all. What they do is mechanically manipulate data and structures of data. All of this is dead, without life, without mind, without perception, without consciousness, and without relation or concern.

What living beings ‘do’ is an entirely other order of activity, and no machine shall ever do the same thing, or even a thing ‘alike’ with what humans do, except in an extremely narrow and superficial form. They can be made by us to »emulate the functionality of human activity; but this is neither learning nor intelligence, and there is no layer of awareness of meaning involved in these emulations.

The reason it is crucial that we understand this is that this concept imposes itself on our ideas and conceptualizations of intelligence and consciousness. If these are collapsed into mechanism, our minds… our humanity… and our intelligence… die. More rapidly the more deep the categorical incursion or forgery becomes.