You think a plane automatically trimming the aircraft out of control range without warning or feedback and keeping it like that while it should had been trimmed back to normal is safe ?
Had you studied the events without any prejudice, you would have noticed that the auto-trim did its job as the maneuver demanded : Back stick and a decreasing airspeed.
At that moment, they went into a stall warning and probably because the system sensed both FACs failed, it went - correctly - into direct law, because of the extended gear.
The crew had at that point two ways of identifying their flight control status :
1/- the lack of the warnings associated with "normal Law" (SPEED, SPEED)
and the lack of automatic TOGA that should have been triggered by an "alpha floor" sensed situation
2/- the change in the PFD and the "Use Man Pitch Trim" message.
As I said earlier, the stickforward movementfrom the PF and the application of manual TOGA was enough to exit tyhe stall situation and had they used the manual trim, it would have been just a classic stall recovery.
With the data we have at our disposal, we can't define the whys of the law changes (logic about failed inputs to the DMCs, comparison with the stby AoiA sensor....)
What we can , IMO, safely surmise is that the failed AoA sensors are at the top of the causes of the crash.
SFLY, there is more to piloting than moving throttles and sidesticks. And your arguments are at least specious. To say that they are what caused a crash is dishonest (just think of the THY 738 in AMS).