As a very experienced Airbus captain, I can assure you that the Airbus is easy to override. Instantly. The automation in an A320 is at a lower level than a B787, and both Boeing and Airbus make manual overrides as simple as move a switch of pushing a button (which in the case of A320 is right by your thumb).
As for your link, every Airbus pilot knows that air show crash was not caused by a failure of the automation, but the foolishness of the man at the controls, depending on an automatic feature he had turned off. All he ever had to do to recover was move the thurst levers like any pilot would.
I can assure you that the Airbus is easy to override. Instantly.
*********************
I’m just a MEI but I’ve been working with computers since the IBM System 360 days... computers in purpose built buildings with stable power and working on problems with a fixed number of variables... the number of variables in a flight system are endless... add to that assumptions that inputs are true and accurate... and external inputs from WX where the actual forces and vectors are truly unknown (or not known with complete accuracy) but assumptions must be made by the flight systems... I have seen “impossible” outcomes too many times in real computer systems to see mobile systems as infallible.
I refer you to the airshow footage ... The pilot demanded TOGA power and the computer said WE’RE LANDING ,, pilot override seems to have FAILED there... then there was AF off Brazil ,, iced pitot and the computer is worthless ,, worse than worthless it’s making control decisions based on bad input and the jet belly flops from FL350.... Then you have the exploding engines issue Airbus has because they certainly can’t use US built engines when they can buy engines from “old Blighty” (RR)...