4 ms·

Nobody is talking about whether it is safe to put Beta software of such an advanced autopilot system in real world? I mean they are giving choice to people to t

Nobody is talking about whether it is safe to put Beta software of such an advanced autopilot system in real world? I mean they are giving choice to people to turn on this Autopilot functionality as if it were just another driver assistance safety feature like 'Honda Sensing' but in reality it is much more advanced system and it seems Tesla is doing this to use the data to feed to it's upcoming autonomous technology. But at what cost! UPDATE: Just read this news on NYTimes: "The traffic safety agency is nearing the release of a new set of guidelines and regulations regarding the testing of self-driving vehicles on public roads. They were expected to be released in July." Good and much needed step from NHTSA here.

wrsh07 10y ago
Tesla doesn't need autopilot to gather data -- just having the sensors on the vehicles is enough. The current cost seems to be about 3/4 as many casualties as one would expect from the # of miles they've driven.
- ceejayoz 10y ago
  > The current cost seems to be about 3/4 as many casualties as one would expect from the # of miles they've driven. That statistic is misleading, though. Tesla's comparing autopilot miles to all driven miles in other vehicles, but they're not directly comparable. Autopilot can only be engaged in certain (favorable) conditions - no residential streets, no undivided highways, and you're probably not turning it on during a thunderstorm or in snow/ice.
  - wrsh07 10y ago
    Sure, it's maybe a bit deflated. My point stands: the cost based on current empirical evidence is negligible. There certainly hasn't been a statistically significant increase in fatal crashes while on autopilot.
ceejayoz 10y ago
I'm very curious about the car insurance implications of using the beta autopilot. I can't imagine Allstate's going to be all that happy paying a $100k claim caused by a product that's explicitly marked as being a beta.
fdsaaf 10y ago
Exactly what standard do you hold beta features before release? For the sake of the future of humanity, we can't get into a situation where nothing new is allowed so long as someone, somewhere might get hurt.
- ceejayoz 10y ago
  > Exactly what standard do you hold beta features before release? With stuff that could easily kill you? If it were a drug, there'd be an ethical review and progressively larger trials in carefully selected groups of people receiving informed consent. Something like that might be more appropriate than "your car just updated! enjoy!" beta software.
  - fdsaaf 10y ago
    You didn't answer the question.
    - ceejayoz 10y ago
      Well, it's going to depend on the feature. Slightly improved UI on the radio's touchscreen is a bit different from Tesla's autopilot. In the case of autopilot, I'd say similar to a drug study, as it's something that can cause serious harm or death to the human subjects involved. Tesla should have an IRB for ethical monitoring, there should be a small trial group initially (and very gradual rollouts to larger groups), participants should receive informed consent at a Tesla store with emphasis on the risks and what's necessary to use the feature safely, and someone should have the authority to pull the plug on the study if it proves too dangerous. Given the state autopilot is reportedly in, I'd say right now no one should be using it without an in-person training at a Tesla store to emphasize the proper usage of it.
      - fdsaaf 10y ago
        Your small-group drug study would have reported zero incidents in this instance.

wrsh07 10y ago

Tesla doesn't need autopilot to gather data -- just having the sensors on the vehicles is enough. The current cost seems to be about 3/4 as many casualties as one would expect from the # of miles they've driven.

ceejayoz 10y ago
> The current cost seems to be about 3/4 as many casualties as one would expect from the # of miles they've driven. That statistic is misleading, though. Tesla's comparing autopilot miles to all driven miles in other vehicles, but they're not directly comparable. Autopilot can only be engaged in certain (favorable) conditions - no residential streets, no undivided highways, and you're probably not turning it on during a thunderstorm or in snow/ice.
- wrsh07 10y ago
  Sure, it's maybe a bit deflated. My point stands: the cost based on current empirical evidence is negligible. There certainly hasn't been a statistically significant increase in fatal crashes while on autopilot.

ceejayoz 10y ago

I'm very curious about the car insurance implications of using the beta autopilot. I can't imagine Allstate's going to be all that happy paying a $100k claim caused by a product that's explicitly marked as being a beta.

fdsaaf 10y ago

Exactly what standard do you hold beta features before release? For the sake of the future of humanity, we can't get into a situation where nothing new is allowed so long as someone, somewhere might get hurt.

ceejayoz 10y ago
> Exactly what standard do you hold beta features before release? With stuff that could easily kill you? If it were a drug, there'd be an ethical review and progressively larger trials in carefully selected groups of people receiving informed consent. Something like that might be more appropriate than "your car just updated! enjoy!" beta software.
- fdsaaf 10y ago
  You didn't answer the question.
  - ceejayoz 10y ago
    Well, it's going to depend on the feature. Slightly improved UI on the radio's touchscreen is a bit different from Tesla's autopilot. In the case of autopilot, I'd say similar to a drug study, as it's something that can cause serious harm or death to the human subjects involved. Tesla should have an IRB for ethical monitoring, there should be a small trial group initially (and very gradual rollouts to larger groups), participants should receive informed consent at a Tesla store with emphasis on the risks and what's necessary to use the feature safely, and someone should have the authority to pull the plug on the study if it proves too dangerous. Given the state autopilot is reportedly in, I'd say right now no one should be using it without an in-person training at a Tesla store to emphasize the proper usage of it.
    - fdsaaf 10y ago
      Your small-group drug study would have reported zero incidents in this instance.