3 ms·

We should expect that, assuming that self-driving cars are in fact safer, insurance rates will go down for people with self-driving cars, and remain the same fo

We should expect that, assuming that self-driving cars are in fact safer, insurance rates will go down for people with self-driving cars, and remain the same for manual drivers.

dwiel 13y ago
The rates for manual drivers will be a function of how safe that pool of drivers are. If a large portion of the manual drivers do it for fun and are more reckless than the current average driver, their rates may in fact go up.
Someone 13y ago
They will go down for manual drivers, too. If you hit your brakes now, you get hit from behind. If the car behind you is a robot, it will come to a stop at 5cm from your car. Unless, of course, the manual drivers counteract by driving more dangerously. Why stop for a red traffic light if all other cars are driven by computer? Can you try and get an autonomous car to drive into a ditch by moving your car very close to it? I think there will have to be laws governing such behavior ('harassment of an autonomous vehicle'), with video cameras in the autonomous vehicles providing the evidence. That also is a reason that I think driverless cars will not work everywhere for quite a number of years. In the typical old city in Western Europe, an automated car will have to negotiate with pedestrians and cyclists for room. Once the pedestrians know that a car will brake if 'challenged', they will change their behavior. I know I would (I guess driving a self-driving car through the center of London or, better yet, Amsterdam in rush hour must be one of the hardest challenges)
- maxerickson 13y ago
  It might work out better than you think in your last paragraph. Auto routed vehicles will reliably choose the 'main' route through an area (instead of whichever they feel is fastest), so segregated routing should work better (and with little additional physical infrastructure you can have one way rush hour streets and such).
  - Someone 13y ago
    In many old city centers, there isn't much room for segregated routing. Also, why would pedestrians or even cyclists care about one way streets? Even if they do, consider the case where streams of cyclists enter a one way street just in front of an automatic car. That could easily happen in Amsterdam (for an impression, see http://bicycledutch.wordpress.com/2013/03/28/amsterdam-bicycle-rush-hour/ http://bicycledutch.wordpress.com/2013/03/28/amsterdam-bicyc... or http://www.nytimes.com/2013/06/21/world/europe/a-sea-of-bikes-swamps-amsterdam-a-city-fond-of-pedaling.html?_r=0 http://www.nytimes.com/2013/06/21/world/europe/a-sea-of-bike...)
    - maxerickson 13y ago
      If there are 3 streets, the pedestrians and cyclists can take the one with no cars on it. In Amsterdam, cars could reliably be routed out and around (I think lots of people probably already use this strategy, but the point is that the auto router would never cheat). Did you understand segregated routing to mean separate streets?

We should expect that, assuming that self-driving cars are in fact safer, insurance rates will go down for people with self-driving cars, and remain the same for manual drivers.

dwiel 13y ago

The rates for manual drivers will be a function of how safe that pool of drivers are. If a large portion of the manual drivers do it for fun and are more reckless than the current average driver, their rates may in fact go up.

Someone 13y ago

They will go down for manual drivers, too. If you hit your brakes now, you get hit from behind. If the car behind you is a robot, it will come to a stop at 5cm from your car. Unless, of course, the manual drivers counteract by driving more dangerously. Why stop for a red traffic light if all other cars are driven by computer? Can you try and get an autonomous car to drive into a ditch by moving your car very close to it? I think there will have to be laws governing such behavior ('harassment of an autonomous vehicle'), with video cameras in the autonomous vehicles providing the evidence. That also is a reason that I think driverless cars will not work everywhere for quite a number of years. In the typical old city in Western Europe, an automated car will have to negotiate with pedestrians and cyclists for room. Once the pedestrians know that a car will brake if 'challenged', they will change their behavior. I know I would (I guess driving a self-driving car through the center of London or, better yet, Amsterdam in rush hour must be one of the hardest challenges)

maxerickson 13y ago
It might work out better than you think in your last paragraph. Auto routed vehicles will reliably choose the 'main' route through an area (instead of whichever they feel is fastest), so segregated routing should work better (and with little additional physical infrastructure you can have one way rush hour streets and such).
- Someone 13y ago
  In many old city centers, there isn't much room for segregated routing. Also, why would pedestrians or even cyclists care about one way streets? Even if they do, consider the case where streams of cyclists enter a one way street just in front of an automatic car. That could easily happen in Amsterdam (for an impression, see http://bicycledutch.wordpress.com/2013/03/28/amsterdam-bicycle-rush-hour/ http://bicycledutch.wordpress.com/2013/03/28/amsterdam-bicyc... or http://www.nytimes.com/2013/06/21/world/europe/a-sea-of-bikes-swamps-amsterdam-a-city-fond-of-pedaling.html?_r=0 http://www.nytimes.com/2013/06/21/world/europe/a-sea-of-bike...)
  - maxerickson 13y ago
    If there are 3 streets, the pedestrians and cyclists can take the one with no cars on it. In Amsterdam, cars could reliably be routed out and around (I think lots of people probably already use this strategy, but the point is that the auto router would never cheat). Did you understand segregated routing to mean separate streets?