Re: Two angle-of-attach sensors not enough
This is where the fact that the crew in the cockpit did not know how to fly became the issue.
This really is the crux of the matter. The computer is very good at following rules, the humans to some extent are there mainly to improvise and make best guesses as to what is actually happening when there isn't enough data for the computer to follow the rules (aided with some additional sensory powers the computer does not have, but hindered by not having other sensory powers the computer does have). The one thing the human can do that the computer cannot is make a series of guided guesses outside the normal flight rules that just might save the airplane from destruction, and even then history shows many humans fail at this fallback task when the computers give up.
In this case the computer absolutely could have kept the airplane flying along, but it would have had to know the problem was only the pitot tubes and that the captain's decision was to treat it as an airspeed sensor failure. At that point it could have kept the wings level and set the engine power to a known safe value. However, Airbus didn't design it that way, Airbus assumed the human backup pilot would take control, do an evaluation, determine the most likely fault, determine the correct course of action for the fault, and keep control of the airplane during and after the failure. The one thing the computer couldn't do was make the decision to handle the fault as a pure airspeed failure -- it just saw disagreeing sensors and, not being intelligent, threw its hands up and said it was done flying.
Simply put, pilots that cannot even keep the shiny side up with a single system inoperative (airspeed) using basic flying techniques are demonstrably worse than the automation in every way and should be barred from anything that looks like a flight deck!