Uber CEO Travis Kalanick says his company’s efforts to automate vehicles is an "existential" decision to be able to survive the next big shift in transportation. Yet when speaking at the Vanity Fair New Establishment Summit in San Francisco this evening, Kalanick says that the technology won’t eviscerate millions of jobs, as many economists and wary onlookers outside Silicon Valley fear. At least, not for a number of years.
"I think people misunderstand how this technology is going to rollout," Kalanick said, responding to a question about Uber’s acquisition in August of self-driving trucking startup Otto. "It’s going to take a while before all those trucks are autonomous everywhere it needs to go." Kalanick mentioned how a freight hauler could be automated to drive on its own only in certain situations, like on long highway routes. "But then it has to back into the distribution center." He added
The idea, at least in the short-term, is that truckers and other drivers would adopt self-driving hardware kits or own a new vehicle with autonomous capabilities, but those drivers would still sit behind the wheel. This is a reasonable view of autonomous vehicle regulation, which won’t necessarily allow people to drive in cars without steering wheels or without their attention on the road for quite some time.
The technology might be quite advanced today from where it was just a few years ago. But there’s a still a long road ahead to determine things like liability in crashes, rules around self-driving in unsafe conditions like snow or rain, and a number of other logistical and engineering challenges that in turn create regulatory complications.
Of course, Kalanick was less vocal about how Uber’s self-driving ambitions will affect its fleet of urban drivers. The company is currently using autonomous cars in a pilot program in the Pittsburgh area, where it’s set up a hardware lab staffed by a considerable amount of former Carnegie Mellon roboticists. Kalanick also confirmed today that Uber has deployed a small number of self-driving cars in San Francisco to collect mapping data.
When speaking to what kind of business model might make sense in 10 or 15 years, when self-driving technology is more fully baked and regulations are in place, Kalanick says it could be a hybrid system. Uber might operate its own self-driving fleet, but it could also purchase or rent vehicles from users. "I think there are a lot of different models for how this is might work," Kalanick said.