4 ms·

All this hang-wringing about the decision a driverless car will make when faced with the trolley problem is pointless. Driverless cars will result in a MASSIVE

All this hang-wringing about the decision a driverless car will make when faced with the trolley problem is pointless. Driverless cars will result in a MASSIVE reduction in deaths of drivers and passengers, a MASSIVE reduction in harm to human pedestrians and non-drivers and to the environment. Driverless cars do not look at their cell phones. Driverless cars do not drive drunk. Driverless cars are not in a hurry, not distracted, not upset by what their boss just said, or by the kids in the back seat. Driverless cars will not err as humans do. Driverless cars will reduce the incidence of "trolley problems" incurred in practice by 2 orders of magnitude or more. Driverless TRUCKS will allow cargo to travel at night, when humans are less likely to be on the road. Which will reduce congestion, which will allow people to get where they're going on time when they ARE on the road. Driverless trucks will KNOW when a car is behind them; there is no "blind spot", ever. And driverless trucks will signal to the car behind them, "you are too close, back off". And it will happen, and no one will get upset about it. Driverless cars will not accelerate pointlessly from a stoplight, only to slam on the brakes a second later, because they did not see the line of traffic ahead. Driverless cars will not engage in petty competition with other driverless cars. Driverless cars will be like Buddhist drivers - steady as she goes, no stress, no drama. That means less fuel consumption, less road wear, less environmental damage. Driverless cars will also give mobility to those who are unable to drive - people with eyesight challenges or physical coordination challenges. Anyone fretting about the ethics of driverless cars along the lines of this article is completely, utterly, radically missing the forest for the trees. Driverless cars will be a huge boon for modern society. (Don't get me started about the idiocy of the statement that "car sales are engineered to bring safety features to the expensive models". Has this author completely missed the lessons of capitalism? )

themartorana 10y ago
This. Every time I hear questioning of driverless cars, this. The moral win is at a macro level, and is so enormous as to render anyone's concerns about nearly-zero-percent-likely scenarios completely uninteresting at best, and actually harmful by slowing the adoption of driverless cars at worst. There is almost zero percent likelihood that the scenarios presented in this article would happen (as pointed out everywhere here). But just for shits and grins, it doesn't matter what decision is made at the macro level. Because while n people may die in a very unlikely scenario, where n is the maximum possible deaths in that scenario, the total number of lives saved in that same moment by driverless cars renders all moral questioning moot. Edit: to boot, driverless cars aren't amoral. The decision tree in these unlikely scenarios have already been programmed in by humans. It's their morality that you find yourself governed by.
Noos 10y ago
Driverless cars will still err, though. They will err like software. The car won't be distracted, but if a software fault causes a google car to classify a certain make of car as invisible due to its lidar signature, or the glitch it causes in the algorithim, it will be far worse. Human error is significant, but software error probably can be more catastrophic because it can affect a huge subset of cars, much more than human error can.
- Ensorceled 10y ago
  Driver error kills about 33K people a year in North America. That's how freaking bad this "software glitch" would need to be.

themartorana 10y ago

This. Every time I hear questioning of driverless cars, this. The moral win is at a macro level, and is so enormous as to render anyone's concerns about nearly-zero-percent-likely scenarios completely uninteresting at best, and actually harmful by slowing the adoption of driverless cars at worst. There is almost zero percent likelihood that the scenarios presented in this article would happen (as pointed out everywhere here). But just for shits and grins, it doesn't matter what decision is made at the macro level. Because while n people may die in a very unlikely scenario, where n is the maximum possible deaths in that scenario, the total number of lives saved in that same moment by driverless cars renders all moral questioning moot. Edit: to boot, driverless cars aren't amoral. The decision tree in these unlikely scenarios have already been programmed in by humans. It's their morality that you find yourself governed by.

Noos 10y ago

Driverless cars will still err, though. They will err like software. The car won't be distracted, but if a software fault causes a google car to classify a certain make of car as invisible due to its lidar signature, or the glitch it causes in the algorithim, it will be far worse. Human error is significant, but software error probably can be more catastrophic because it can affect a huge subset of cars, much more than human error can.

Ensorceled 10y ago
Driver error kills about 33K people a year in North America. That's how freaking bad this "software glitch" would need to be.