[LINK] Robot cars and the fear gap

[Brendan]

On 07/14/2016 11:36 AM, Jim Birch wrote:
> Brendan wrote:
>
> It's not clear to me that overall lower injuries/fatalities overrides the
>> manner in which they're suffered/ who they're suffered by.
>
>
> This sounds a little spooky to me.  Are you saying you don't mind being
> injured or killed provided there is a good redeeming narrative available? :)
>
> Flying is relatively very safe but plenty of people find it scary. That
> doesn't seem to me a valid reason for banning airlines or something.
> Seatbelts were considered to have a risk of trapping occupants in an
> imperiled vehicle.  They have produced an immense reduction in RTA death,
> injury and economic loss.  Maybe someone was actually trapped in burning
> car by their seatbelt sometime.  This might be a cause for a seat belt
> redesign but it certainly is not a reason to abandon seatbelts.  Self
> driving cars potentially offer a similar jump in safety.
>
> Can you give an example?

Yes. It's effectively the trolley problem. Do you throw a fat person in the way of a runaway trolley in order to save 5 other people?
Most people say they wouldn't, even though the number of people killed is substantially lower (1 v 5).

Presumably, driverless cars are going to disproportionately remove drunks, suicides and young men from the accident statistics. If there is only a marginal improvement in _overall_ statistics, then that implies that they're being balanced by losses from other groups, so you are effectively choosing who will be killed on the roads.