The problem really seems to be lack of a graceful failure mode of the pitot system and the potential severity of pitot failures on a fly-by-wire aircraft. THAT is the problem will yield best results to an engineering fix. Pitot tubes need to work, but most of the discussion regarding fixing pitot tube problems is really a discussion on the best way to put a band aid on the problem.
The computers that use airspeed need to have a speed input that cannot be completely corrupted by failure of any one class of input. Loss of pitot data should be substituted with SatNav or Inertial speed data in a logical voting progression. Then when pitot data failed, a realistic substitute piece of information could be inserted. This is an engineering problem that engineers can solve. Not cheap ultimately, but compared to the AF447 accident cumulative costs, reasonable.
Sid