Rise of AI

I can’t see AI controlling the human race in the near future. Though it’s certain that incredible AI system exists that far outstrip certain human qualities, there is no AI that exists that could create its own qualities, i.e., the AI of today can’t decide not to do a task or change a task to something completely different by its own volition.

The free will of AI simply doesn’t exist in any form. Currently an AI program executes a learning/reward style which basically tells the program what the outcome should be. If you want your AI to play a computer game, you’d tell your program to complete the level over and over again and each time try and complete it better.
That said, your program has access to the controls that operate the game, it may make the character jump repeatedly, it might walk backwards and forwards ... it being a program all it is registering is that each time it tries to play it dies and so must execute a different move. It doesn’t really know its playing a game, its just running a new program each time it plays ... then when it dies on the game, the old program is terminated and a new program is created to test the next try. The new program have 1 tiny change made.

To see this in action would convince you it’s a super learning machine, but it isn’t, its probably played about 3000 games before it completed it without dying. Just let the program play a new game ... and it’s back to square one. 1000 attempts later and it might know how to play two games, but the chances of it completing a game its never seen in 3 or less attempts is in my opinion ... not possible.
AI can’t freely decide outside of the boundaries we set ... that’s free will.

The rise of the machines is still some way of ...