For as long as I have been hearing about autonomous vehicles, I have been reading about potential lives saved if humans were no longer driving. 37,000 Americans died in car crashes last year and, in 2013, according to the World Health Organization, 1.25 million people died. These numbers seem high but given the amount of car usage, car safety is relatively high and has been improving markedly of the years. While autonomous vehicles may improve these numbers, if we are trying to allocate resources to save lives, this has to be a low priority.
Given the title of this post, that conclusion may surprise you. But the reason I pull away from that metric — lives saved — is that it obscures a more important reality: we already have and continue to spend a ridiculous amount of time on car safety. In other words, we are paying to keep that number as low as it is.
Consider, for a moment, the notion that human error could be completely eliminated by having autonomous vehicles. Would we need car airbags? Would we need seat belts? Would we need the car to be reinforced to withstand high impact collisions? Would we need crumple zones? I could go on. Would we need road barriers? Would we need steet lights? Now think about the costs of all of those things and you can see how they add up.
In the New York Times this week, David Leonhardt wrote an entire piece about getting comfortable with autonomous driving — as if that is the main issue. And he is an economist. His nervousness was not being in control. Why? Because the roads are full of human drivers. That is the reason we want control. Move everyone to autonomy and that anxiety — that is repressed in each of us — goes away. Now that could be a considerable saving in ‘utility costs.’
In the future, as many before me have said, we will laugh at the time we let anyone drive. We will barely understand it. But even more we won’t be able to fathom the cost of accommodating human control.
3 Replies to “The Safety Cost Savings for Autonomous Vehicles are Way Understated”
Anybody thinking autonomous vehicles eliminate all auto deaths is probably ripe to buy a bridge in Brooklyn. There are accidents which a perfect driver could not avoid (e.g. ice in winter, I’ve spun off a road at low speed). If safety features are removed from autonomous cars we’ll need to shut the car down during conditions the car can’t guaranty.
Which adds a new excuse: “Please excuse Timmy from school, the car refused to leave the house this morning.”
@ Thomas Wicklund
I notice that you didn’t die when you spun off the road in icy conditions. Very, very few fatalities occur at low speed. You are probably technically correct that auto deaths wont drop to zero, but dropping to a tiny fraction of the current amount is definitely a possibility.