The future is coming

Yes. They don't work because it would eventually result in the robot entering and endless cycle of yes/no. Like in Wargames the movie when the computer plays tic tac toe.

No, not as all. If true AI robots are programmed to respect all FOUR of Asimov's laws, there is a clear priority and no conflict. This robot is not respecting the 1st law, probably due to its following some crappily designed code full of human errors. We haven't achieved full, Asimovian AI yet :-)

There are only FOUR laws because the original three didn't work.