We need not fear Google’s self-driving car. But perhaps the self-driving car should fear us.
In several years of testing, Google’s autonomous cars could certainly qualify for a safe-driver insurance discount. They always do the right thing, like a youth who just passed a permit test and remembers all those fussy little rules merged with a computer that betrays no emotion. But these cars seem to be such sticklers for the law that they have had a hard time dealing with human drivers who don’t worry quite so much about these things.
In Google’s latest accident report, released on Tuesday, one incident highlights that conundrum: While nearing a crosswalk, the Google car slowed. The “safety driver” inside (these cars are a long way from being allowed to go about town on their own) reasonably applied the brakes as well. The pedestrian was fine, but the car was hit from behind by another car that couldn’t stop in time.
Interestingly, a Google examination of the accident determined that if the car had been left on its own, it would have stopped closer to the crosswalk, giving the car behind it more time to stop. Would it have been enough to avoid the accident? That’s hard to say, but it lends credibility to the notion that the roads would be a lot safer if humans were removed from the equation.
But that’s still science fiction. For now, autonomous cars — or their programmers — will have to learn to deal with angry bike messengers, people who drive too close, distracted drivers and the other obstacles of an imperfect and very human world.