Nearly 40,000 people will die in car accidents this year in the United States. But so far, only one of those people has died behind the wheel of a self-driving car. Yesterday, news broke about a 40-year-old man name Joshua Brown who was killed after his Tesla Model S — operating in the car’s beta “autopilot” mode — collided with a tractor-trailer in Florida earlier this spring. The trailer made a left turn and the Tesla’s brakes did not engage. Tesla says that “neither Autopilot nor the driver noticed the white side of the tractor trailer against a brightly lit sky,” so the brake was not applied.
Brown, in the words of Tesla’s press release, “believed strongly in Tesla’s mission.” Before his death, he posted a YouTube video showing his Tesla swerving to avoid collision with a truck on the road. “I actually wasn’t watching that direction and Tessy (the name of my car) was on duty with autopilot engaged. I became aware of the danger when Tessy alerted me with the ‘immediately take over’ warning chime and the car swerving to the right to avoid the side collision,” he explained in the video’s caption.
But the technology isn’t perfect. According to the company, every Tesla “disables Autopilot by default and requires explicit acknowledgement that the system is new technology and still in a public beta phase before it can be enabled.” Tesla technology is known to have problems identifying tractor-trailers on the road and it’s possible that Brown also didn’t see the other vehicle in time, either. (The National Highway Traffic Safety Administration is conducting a formal report on what happened in Florida.) But either way, Brown’s death is likely not going to be an isolated incident.
Self-driving cars conjure up images of a future in which everyone is a passenger. Drunk? Hop in and let your car safely get you home. Tired? Take a nap while your car minds the road. When Elon Musk introduced the Model S (the same model in which Brown died) he called the vehicle’s autopilot “probably better than a person right now.” This carries with it enormous potential safety benefits. Tesla says that there is one traditional car fatality for every 94 million miles driven, while Brown’s death is the first after 130 million miles driven in Teslas on autopilot.
But despite Musk’s glowing praise of his own product, the world isn’t quite there yet, because self-driving vehicles remain costly (an entry-level Tesla will run over $35,000) and because autopilot tech actually hasn’t evolved to that point yet. Even when autopilot is engaged, Tesla recommends drivers keep their hands on the wheel at all times. Instead, the roads remain a hodgepodge of traditional and smart cars. As Brown’s crash showed, this can cause major issues.
Traditional car deaths, while always tragic, have become a relative norm. We know that whether we’re getting into Honda Civic or a Ford F-150 that driving is inherently dangerous. When people die in auto accidents, it’s shocking but not surprising. But humans have an inherent skepticism about new technologies and tend to, at the same time, overestimate their own abilities. No matter how safe you can be assured a self-driving car might be, a lifetime of being behind the wheel will lead you to believe that you have a better sense of the road than a car does. There’s a philosophical element to it, maybe — better to be in charge of your own destiny than leave it to a robot, even if the robot is technically better than you.
Every death in a self-driving car will reinforce this feeling. And there will be others. Self-driving technology has the aura of inevitability, and while it exists in an abstract future, it feels like a wonderful dream. But as it moves from early adoption to wider use, and from “beta” to a less-hedged official feature, we’re going to have to confront the fact that “safer” doesn’t mean “perfect.”