Main question for me is, why doesn't it use gyro meters for figuring out a rough approximation of AoA? So it would only kick in if gyro is above a certain bound

Main question for me is, why doesn't it use gyro meters for figuring out a rough approximation of AoA? So it would only kick in if gyro is above a certain boundaryb and the air speed sensor is below threshold. Then double up both the gyros and the air speed sensors, if anything disagrees give errors instead of activating. Even outside of aviation I find it incredible how naive the current design is.