Aren't we now approaching a tower that isn't "designed to fail" at all, then? The twins themselves can hardly be said to have been designed to fail, and yet both failed with spectacular speed and destructive power from two very different initiation points. ...
The difference is quantitative, not qualitative.
It is often said that the twins, like other highrises, were designed with some factor of safety - or their structural elements. The idea being that the elements could bear, say, 2-3 times their maximum load; or the assembly 2 times the weight on top (I am making these number up, but you recognize the concept).
(Reality is more complex than this, but you get the point)
You could turn this on its head and say that "
the twins were designed to fail at 2 times their max load", or "
the twins were designed to fail at or below 50% of capacity". Not literally, not consciously, but in effect.
This "factor of safety" is maintained only for the as-designed assembly.
Once the assembly has taken wide-spread fire damage, once it has suffered structural damage from plane crashes or debris smashing into the walls, they are no longer "as-designed", and no factor of safety can be assumed any longer.
Once the top part has started to move downward more than seismic stress could induce, you are outside the envelop - no factor of safety can be designed into the structure to accomodate this mode of being. It's a case that a building simply not "designed for". The progressive failure and dynamic loading both increase local loads and decrease local capacity.