3 ms·

Is this how it works with medicines? When you say that things can still happen, you mean that people will die - right?

maxxxxx 8y ago
I am talking about medical devices but I assume it's the same everywhere. You demonstrate that you have done everything possible to analyze the risks and taken appropriate steps. But there is always something risk left that can't be foreseen. It's like building an airplane. If you follow all best practices the plane will be pretty safe but from time to time some still will crash. There is almost no way around that.
- sorokod 8y ago
  Not familiar with airplane building, if you take away the human factor as is the case with self driving car, what risks would would be considered acceptable?
  - maxxxxx 8y ago
    No idea. But it's pretty much guaranteed that autonomous cars will kill some people due to bugs and other problems. I assume there will be regulations like we always have for crash tests which manufacturers will have to follow.
    - sorokod 8y ago
      Right, something like "if you have x% test coverage and use Go, you are not liable for accidents"?
  - ksdale 8y ago
    It seems to me like as soon as you can demonstrate that the cars are safer than the average driver, they should be okayed for road use. An absolutely staggering number of people die in car crashes. People will of course die in and around autonomous vehicles as well, but the key metric is whether autonomous vehicles cause fewer deaths than the average driver. Similarly, you could have a late stage cancer drug that kills half the people who take it and saves the other half and it provides a massive benefit if 100% if all those people were going to die otherwise. The fact that a lot of people die is not the only factor to be considered. What matters is how the ultimate outcomes are changed. It would be quite irresponsible to say that the cancer drug shouldn't be allowed solely because it kills half the people who take it.
    - sorokod 8y ago
      This still leaves the question of liability. In the case of cancer treatment I assume that patients are required to sign some sort of waver. Don't see how this can work in the self driving car case as some of the victims may not have signed such a waver. This leaves some sort of bulk protection for manufacturers ?
      - ksdale 8y ago
        I don't know what will actually happen but I'm partial to strict liability for manufacturers. If the cars actually cause fewer accidents than regular cars, then even if manufacturers are liable for every accident, the total cost would be less than what people currently spend on car insurance. The cost would be initially paid by manufacturers and I'm assuming passed on to drivers through the purchase price, resulting in basically the system we have now, without the drivers of driverless cars having to carry their own insurance... Sounds good in theory, anyway

I am talking about medical devices but I assume it's the same everywhere. You demonstrate that you have done everything possible to analyze the risks and taken appropriate steps. But there is always something risk left that can't be foreseen. It's like building an airplane. If you follow all best practices the plane will be pretty safe but from time to time some still will crash. There is almost no way around that.

sorokod 8y ago
Not familiar with airplane building, if you take away the human factor as is the case with self driving car, what risks would would be considered acceptable?
- maxxxxx 8y ago
  No idea. But it's pretty much guaranteed that autonomous cars will kill some people due to bugs and other problems. I assume there will be regulations like we always have for crash tests which manufacturers will have to follow.
  - sorokod 8y ago
    Right, something like "if you have x% test coverage and use Go, you are not liable for accidents"?
- ksdale 8y ago
  It seems to me like as soon as you can demonstrate that the cars are safer than the average driver, they should be okayed for road use. An absolutely staggering number of people die in car crashes. People will of course die in and around autonomous vehicles as well, but the key metric is whether autonomous vehicles cause fewer deaths than the average driver. Similarly, you could have a late stage cancer drug that kills half the people who take it and saves the other half and it provides a massive benefit if 100% if all those people were going to die otherwise. The fact that a lot of people die is not the only factor to be considered. What matters is how the ultimate outcomes are changed. It would be quite irresponsible to say that the cancer drug shouldn't be allowed solely because it kills half the people who take it.
  - sorokod 8y ago
    This still leaves the question of liability. In the case of cancer treatment I assume that patients are required to sign some sort of waver. Don't see how this can work in the self driving car case as some of the victims may not have signed such a waver. This leaves some sort of bulk protection for manufacturers ?
    - ksdale 8y ago
      I don't know what will actually happen but I'm partial to strict liability for manufacturers. If the cars actually cause fewer accidents than regular cars, then even if manufacturers are liable for every accident, the total cost would be less than what people currently spend on car insurance. The cost would be initially paid by manufacturers and I'm assuming passed on to drivers through the purchase price, resulting in basically the system we have now, without the drivers of driverless cars having to carry their own insurance... Sounds good in theory, anyway