The Angle of attack sensor - possible bad readings. I am concerned that these have not been sampled and tested - as the urgency appears to be real. the sensor is the sharp end, and fools are being tricked into look at the software. There are many ways to make an AOA sensor. Airbus uses a static tube and holes - I think.
Expensive complicated, andif power to the heater fails, very obvious.. Boeing uses a vane - could be (potentimeter RVDT/linear variable differential transformer (LVDT)/rotary encoding What is the 737 Max one?
Link: Angle of Attack Vane - Basic Air Data
Potentiometer - because we know volume controls can get contaminated. salt, solvents, cleaning fluid, detergent or because the leads carry a low voltage - so connectors can be dirty as well.
US5438865A - Angle of attack sensor - Google Patents
swept back design of the vane will discourage service personnel from using the vane as a handhold or foot rest
Rotary ones can osiillate , wind gusts, turbulence.Ball bearings may sieze if enough desolvet hits it.
I reckon it is probable that Boeings is not averaged, and is of the volume control design, software reading every 1/2 second or so, and someone is working out an envelope is need to eliminate out of scope readings.
A neat solution would be keep one B encoder, and fit an airbus/tube aoa on the other side. Much more failsafe.
Technically a sanity AOA could just be an angled hole and an anemometer behind it adding its 2 cents worth on an I2C bus.