[Stubbs]

and how many times has AP avoided accidents the driver wouldn't? The stats arn't that simple any more are they.

There are plenty more news articles, especially in local news, about fatalities involving non autopilot cars, but they're not as interesting so don't get shared as often.

[close04]

Such stats are never simple. But when important data is withheld they definitely become irrelevant. They become more of a PR tool.

Statistically speaking 100% of people are able to pilot a plane with their eyes closed and not crash it... as long as a qualified pilot is ready to take over at any time. Such a stat is only valuable if you want to show the standby pilot is useful.

What would you base your assumption that AP saved even 1 life on when that data is not provided? All we know is that it crashed a number of times exclusively when it wasn’t supervised as required. The pragmatic conclusion is that it works because it’s supervised. I imagine Tesla would release more data if it supported the performance of the AP.

[mrep]

Here is a cool video in which shows a model 3 dodging a car in front of it [0] after it got rear ended causing it to jut forward. Owner says it wasn't him but autopilot that did it.

[0]: https://www.reddit.com/r/teslamotors/comments/bl5pa9/tesla_m...