2 ms·

Would we say the equivalent thing about the ill-fated MCAS in the Boeing 737 MAX? How different is TSLA's behavior compared to what Boeing's position was when i

by vivegi 4y ago

Would we say the equivalent thing about the ill-fated MCAS in the Boeing 737 MAX? How different is TSLA's behavior compared to what Boeing's position was when it didn't inform the pilots about how MCAS worked (and actively called it safe) until it was forced to do so by the regulator's investigation into the crashes. Basically what TSLA is saying is -- Our all-knowing AI is infalliable. Stupid drivers should just keep their hands off and they don't need to know. The easiest thing to do would be to tar the messenger and ignore the message. Didn't the TSLA autopilot a few years ago have problems identifying a slow moving tractor trailer and a parked fire truck. https://arstechnica.com/cars/2019/05/feds-autopilot-was-active-during-deadly-march-tesla-crash/ https://arstechnica.com/cars/2019/05/feds-autopilot-was-acti... https://www.caranddriver.com/news/a30474504/nhtsa-indiana-tesla-model-3-crash-fire-truck/ https://www.caranddriver.com/news/a30474504/nhtsa-indiana-te... I guess we should just ignore all evidence to the contrary and drink the TSLA koolaid on FSD/autopilot.

mlindner 4y ago
> Would we say the equivalent thing about the ill-fated MCAS in the Boeing 737 MAX? Completely apples and oranges though? The pilots on the MAX weren't intentionally trying to defeat a safety feature.
- salawat 4y ago
  Actually, thy were, because said safety feature was malfunctioning. The definition of malfunctioning with anything based on Machine Learning however, is much murkier, because there is no guarantee that the function you've learned is actually the one you think you've learned.