5 ms·

This is probably a large underestimate given that Tesla tends to be very aggressive on not sharing all their data and not calling things autopilot related when

This is probably a large underestimate given that Tesla tends to be very aggressive on not sharing all their data and not calling things autopilot related when they probably should be. Than being said, even if we 10x this I would guess that number is still pretty low. The comparison should not be relative to 0, the comparison should be relative to number of equivalent human driving hours. How many teslas on the road are out there, how many hours/miles have they driven, and how does that stack up to average human driving in comparison? Obviously those stats can be gamed, but I'm generally inclined to say that as long as there isn't some crazy increase then its a step in the right direction.

bmcahren 3y ago
> We also receive a crash alert anytime a crash is reported to us from the fleet, which may include data about whether Autopilot was active at the time of impact. > To ensure our statistics are conservative, we count any crash in which Autopilot was deactivated within 5 seconds before impact, and we count all crashes in which the incident alert indicated an airbag or other active restraint deployed. (Our crash statistics are not based on sample data sets or estimates.) > In practice, this correlates to nearly any crash at about 12 mph (20 kph) or above, depending on the crash forces generated. We do not differentiate based on the type of crash or fault (For example, more than 35% of all Autopilot crashes occur when the Tesla vehicle is rear-ended by another vehicle). In this way, we are confident that the statistics we share unquestionably show the benefits of Autopilot. Every answer to your questions is on the link below: Source: https://www.tesla.com/VehicleSafetyReport https://www.tesla.com/VehicleSafetyReport
- BoorishBears 3y ago
  The Tesla safety report has never been a good-faith attempt at communicating their stats. They compare Teslas to every single other vehicle on the road. The average Tesla owner is already in a demographic that would have dramatically lower accident rates than the average. The average age of a Tesla vehicle also dramatically lowers it. Then you add in the fact AP is skewed towards highway use and disengages in tricky situations... Before you even factor in AP you're already starting with the picture Tesla wants to paint.
taylodl 3y ago
> the comparison should be relative to number of equivalent human driving hours Not necessarily, this would be true if autopilot made the same kinds of mistakes that humans do - these are the kinds of mistakes that other humans, e.g. other drivers, motorcyclists, bicyclists, pedestrians, are on the lookout for. That is to say we're very experienced with the kinds of mistakes humans make and the conditions in which they make them, and so we can defend ourselves against them. If autopilot is making all-new mistakes under totally different conditions then it removes that ability to defend ones self and lead to even more issues. We humans have a keen sense for how other humans will react in a given situation.
- dgrin91 3y ago
  Do we have the ability to defend ourselves? How do you defend against a drunk driver? Lots of people die to random car crashes because they aren't able to defend themselves. The promising thing I've always thought about self driving is that while it may make stupid, unpredictable mistakes (drive into flashing firetruck at 60mph) it does this less than a typically human driver would make typical human casualty inflicting mistakes. Moreover we can build & improve to eliminate firetruck ramming; we unfortunately have not been able to eliminate drunk driving.
  - taylodl 3y ago
    We've been unwilling to eliminate drunk driving. Breathalyzer ignition lock? Problem solved. Self-driving cars are an extremely poor solution (complexity and cost) to solve this simple problem. People have fought tooth and nail to enable drunk driving - and they've won. Just like they're fighting and winning for mass shootings. Even so, drunk drivers are largely predictable. People know when it's not a good time to be on the streets.

bmcahren 3y ago

> We also receive a crash alert anytime a crash is reported to us from the fleet, which may include data about whether Autopilot was active at the time of impact. > To ensure our statistics are conservative, we count any crash in which Autopilot was deactivated within 5 seconds before impact, and we count all crashes in which the incident alert indicated an airbag or other active restraint deployed. (Our crash statistics are not based on sample data sets or estimates.) > In practice, this correlates to nearly any crash at about 12 mph (20 kph) or above, depending on the crash forces generated. We do not differentiate based on the type of crash or fault (For example, more than 35% of all Autopilot crashes occur when the Tesla vehicle is rear-ended by another vehicle). In this way, we are confident that the statistics we share unquestionably show the benefits of Autopilot. Every answer to your questions is on the link below: Source: https://www.tesla.com/VehicleSafetyReport https://www.tesla.com/VehicleSafetyReport

BoorishBears 3y ago
The Tesla safety report has never been a good-faith attempt at communicating their stats. They compare Teslas to every single other vehicle on the road. The average Tesla owner is already in a demographic that would have dramatically lower accident rates than the average. The average age of a Tesla vehicle also dramatically lowers it. Then you add in the fact AP is skewed towards highway use and disengages in tricky situations... Before you even factor in AP you're already starting with the picture Tesla wants to paint.

taylodl 3y ago

> the comparison should be relative to number of equivalent human driving hours Not necessarily, this would be true if autopilot made the same kinds of mistakes that humans do - these are the kinds of mistakes that other humans, e.g. other drivers, motorcyclists, bicyclists, pedestrians, are on the lookout for. That is to say we're very experienced with the kinds of mistakes humans make and the conditions in which they make them, and so we can defend ourselves against them. If autopilot is making all-new mistakes under totally different conditions then it removes that ability to defend ones self and lead to even more issues. We humans have a keen sense for how other humans will react in a given situation.

dgrin91 3y ago
Do we have the ability to defend ourselves? How do you defend against a drunk driver? Lots of people die to random car crashes because they aren't able to defend themselves. The promising thing I've always thought about self driving is that while it may make stupid, unpredictable mistakes (drive into flashing firetruck at 60mph) it does this less than a typically human driver would make typical human casualty inflicting mistakes. Moreover we can build & improve to eliminate firetruck ramming; we unfortunately have not been able to eliminate drunk driving.
- taylodl 3y ago
  We've been unwilling to eliminate drunk driving. Breathalyzer ignition lock? Problem solved. Self-driving cars are an extremely poor solution (complexity and cost) to solve this simple problem. People have fought tooth and nail to enable drunk driving - and they've won. Just like they're fighting and winning for mass shootings. Even so, drunk drivers are largely predictable. People know when it's not a good time to be on the streets.