Uber Self-Driving Car Kills Arizona Woman, Realizing Worst Fears Of The New Tech
By Marco della Cava, USA TODAY
March 19, 2018
SAN FRANCISCO — An Uber self-driving car hit and killed a woman in Tempe, Ariz., Sunday night, an incident that might deliver a setback to the otherwise feverish development of autonomous transportation.
The Uber car was in self-driving mode when the accident occurred, but had a safety driver at the wheel as is the norm when testing such vehicles, according to local police.
Uber issued a statement saying that it was “fully cooperating with local authorities.” In wake of the incident, the ride-hailing company has temporarily halted its self-driving testing in Pittsburgh, San Francisco and Toronto, as is its policy following any accident.
The woman, who has not yet been named, was crossing outside the designated crosswalk at about 10 p.m. when she was hit, police said.
Sgt. Ronald Elcock, a Tempe police spokesman, confirmed to USA TODAY Network that the car was in autonomous mode with a driver behind the wheel when it hit the pedestrian. There were no passengers in the car at the time.
In a tweet, National Transportation Safety Board officials said they were sending a team to Arizona to investigate the accident.
Federal regulators have been supportive of the move toward autonomy, with transportation secretaries under both Obama and President Trump creating a framework for companies to share developments without giving away trade secrets.
Uber CEO Dara Khosrowshahi tweeted Monday that the company had “some incredibly sad news out of Arizona. We’re thinking of the victim’s family as we work with local law enforcement to understand what happened.”
This appears to be the first known pedestrian death involving a self-driving car, a technology that has long been touted by its supporters as being able to drastically reduce the number of automobile fatalities.
Around 37,000 were killed in traffic accidents in 2016, up 6% from the previous year, according to the National Highway Traffic Safety Administration.
The incident is sure to add fuel to concerns on the part of some activists, such as Consumer Watchdog, who feel the technology that allows the cars to sense their surroundings is not ready for prime time.
Critics claim that legislators in autonomous-friendly states such as California, Arizona and Michigan are too eager to accommodate such testing in the hopes of being early winners in a new mobility sweepstakes.
“There should be a national moratorium on all robot car testing on public roads until the complete details of this tragedy are made public and are analyzed by outside experts so we understand what went so terribly wrong,” says John Simpson, privacy and technology project director at Consumer Watchdog.
“Arizona has been the wild west of robot car testing with virtually no regulations in place,” he says. “That’s why Uber and Waymo test there. When there’s no sheriff in town, people get killed.”
Uber and Waymo, the name of Google’s self-driving car company, have been testing self-driving cars aggressively in the Phoenix area for a few years.
Waymo has been picking up beta-program passengers in its self-driving Pacifica minivans since last year, and later this year plans to open up the program to a wide Phoenix audience.
The company also just started operating without a safety driver overseeing the ride from the passenger seat.
It remains too early to tell if the fatality will slow the feverish pace of self-driving car progress. Large automakers and technology companies alike have been accelerating their efforts to develop commercially viable self-driving cars, which promise a host of benefits.
Paramount is their ability to cut down on the soaring number of traffic deaths, which are partly due to distracted driving in the age of the smartphone.
But self-driving cars, which run on electricity, also would be a boon to ride-hailing companies such as Uber and Lyft, whose business models currently are hamstrung by having to pay drivers.
Automakers are keen to embrace self-driving fleets in order find a new revenue source as new generations of would-be drivers opt for transportation-as-a-service options over traditional and costly car ownership.
But questions abound about self-driving cars, ranging from insurance to their ability to deal with inclement weather. But perhaps the biggest unanswered questions — one that might be pushed to the fore in the wake of the death in Arizona, — is whether consumers ultimately will prove comfortable being shuttled by a robot.
So far, public views on autonomy could be sobering for companies pumping billions into self-driving cars.
A 2017 survey by managment consultants Alix Partners revealed that 49% of consumers said they don’t currently feel confident in AVs’ abilities to navigate them safely, with 84% of respondents saying they were concerned about vehicle-software malfunctions and 80% saying the same about potential hardware malfunctions.
Ultimately, we may be in a difficult transition period where artificial intelligence systems such as those that power self-driving cars aren’t sophisticated enough to deal with every situation, says Timothy Carone, an associate teaching professor in the Department of IT, Analytics, and Operations at the University of Notre Dame’s Mendoza College of Business.
“Transitioning the knowledge that human beings have acquired over the decades of driving and flying over to autonomous systems like a driverless car or unmanned plane is a complex undertaking that will itself takes years of work for society to recognize the benefits of autonomous systems,” he says, adding that what follows is a painful period where these systems improve gradually while human skills diminish.
Says Carone: “There will be a time when the autonomous system is not dealing correctly with a problem yet the humans will have lost their expertise and will not be able to take over from the autonomous system to prevent a tragedy.”
Contributing: Gabrielle Olivera and Ryan Randazzo with The Arizona Republic.
Follow USA TODAY tech writer Marco della Cava on Twitter.