That's interesting though I suppose unsurprising. We did a PoC with Infineon, and demoed at CES this last January, something related to the limitations of Fail

That's interesting though I suppose unsurprising. We did a PoC with Infineon, and demoed at CES this last January, something related to the limitations of Fail-Safe mechanisms in autonomous systems and what changes in deisgn & runtime context are essential for moving from Fail-Safe to Fault-Tolerant systems. In the case of MCAS this probably works out okay because the remediation is to disable and hand control back to the human (something that I'm frankly shocked wasn't already part of the hazard and risk assessment) in loop. Alternative design approaches could probably extend the safe operational envelope for MCAS to tolerate more (or simultaneous) dimensions of failure, but outside of making MCAS a more valuable feature to market (which isn't that useful since it's baseline essential for the aircraft to even be certifiable) it's easy to see the calculation for not going down that redesign road.