I believe that most pilots would agree that the correct action to take to recover from a stall is not intuitive or instinctive. It has to be learned. So we send up students and instructors, where one of them has a demonstrated skill set, and the other doesn't, and we leave it to which one is physically stronger to determine the outcome of a bad situation. To make if worse, the panicked student likely may be all wound up on adrenaline, which doesn't help the instructor at all. The FAA and others have spend billions, collectively, on safety improvements, most recently with all sorts of electronic/computer tech, but we're still using the same mechanical coupling for student and instructor that was being used 100 years ago.
I know that we aren't losing hundreds of people a day in instruction, as we are with automobile accidents, but it still seems silly to me that there's not a solution to this.
In the RC world, with the radios I used to fly with, there was a lever on the left that you could use to give the student the ability to control the aircraft. To take over, the instructor just lets go of that lever, and control belongs to the instructor. All of this to keep a student from flying the model plane into a bystander, or stall and crash it. And yes, way, way simpler in a system that is already all electric and doesn't have to be certified at all. How many full sized training accidents a year would have been recoverable if the instructor had an ability to absolutely override the student? This thread's accident is one of them, pretty clearly, in my opinion.
Until then, we just say "Yep, statistically these flight will be fine. As long as your son/daughter doesn't panic and kill them both, everything will be fine. And that's pretty rare. Only happens a couple of times a year. Not really worth fixing. Way bigger problems in the world to worry about than saving only X lives a year. Plus, it would be hard to do."
Ok, I'm rambling/ranting again. Soapbox down.