Ahhhhhh..... back to the control law logic and implementation.
From Doze:
It won't do anything unless the pilot asks it to though!
I don't think that's entirely accurate.
Best I can tell from the reversion sequence, the "autotrim" for the THS is in effect until in DIRECT law ( or maybe sub-law 2 (c) iv /sarc off). So the system will attempt to achieve 1 gee corrected for pitch attitude. Have I read that wrong? And I also see an implementation that trims the THS in order to reduce the stick inputs required to maintain the gee required to maintain the previously commanded pitch attitude. NOT A PURE ATTITUDE HOLD mode many of us were used to in A/P -assisted modes.
So relaxing stick with a positive pitch attitude would "normally" result in increasing THS until the AoA functions kick in. Their "mode" didn't have the AoA "protections", if I read that law correctly.
I must agree that manually flying a basic jet at high altitude and a mach near 0.85M is not nearly as benign as the same IAS at 15,000 or 20,000 feet. That mach doofer causes some strange things, and your TAS is much higher so you muct be gentle and use measured inputs. So being a low altitude glider pilot may not satisfy the criteria.
Still a sad, sad example of airmanship and systems knowledge and crew coordination.