SAN FRANCISCO – Self-driving car advocates and observers are reacting with cautious approval Tuesday to the government's 112-page directive on the transformational technology.
"The devil is in the details, so we will want to take a good hard look before we comment," says David Strickland, general counsel for the Self-Driving Coalition for Safer Streets, which advocates for Ford, Google, Uber, Lyft and Volvo. Strickland is also a former administrator of the National Highway Traffic Safety Administration.
"We see this document as evolutionary," he told reporters on a conference call. "But we appreciate their effort, and the iterative process."
Michael Harley, analyst for Kelley Blue Book, adds that the new guidelines "should help standardize the unregulated Wild West chaos that the rapidly advancing field of autonomous technology now operates under."
The result should be autonomous vehicles that deliver a high degree of passenger and pedestrian confidence, he says.
"Too often, the law is behind the technology, and as it stands today, the states have a patchwork quilt of regulations some of which are inconsistent with the concept of autonomous vehicles," notes Sharon Klein, chair of the privacy, security and data protection practice at Silicon Valley law firm Pepper Hamilton. "Today’s guidance is a great step to bridge the gaps, begin educating all constituents and get out in front of the autonomous vehicle trend which will be a game changer for everyone driving."
Even perennial industry skeptic John Simpson of the non-profit group Consumer Watchdog hailed the policy paper.
“This isn’t the checkered flag to industry to irresponsibly develop robot cars that we had feared," Simpson says. "It’s not a secret, cozy process with the manufacturers, but includes a real commitment to transparency and public involvement. The administration clearly heard the concerns raised by safety advocates and has addressed many of them.”
The government's treatise lands as self-driving car companies and road-tests are beginning to mushroom. In the past weeks alone, Ford announced it would build a driverless and steering wheel-less vehicle for ride sharing by 2021, Uber began picking up Pittsburgh passengers in self-driving cars, and Lyft cofounder John Zimmer posted an essay predicting that in five years Lyft's fleet would be largely autonomous.
The main thrust of the Department of Transportation's delayed Federal Automated Vehicles Policy, initially slated for release a few months ago, is to alert automakers and technology companies working on self-driving cars that the department intends to proactively seek access to evolving software and hardware developments in the name of safety.
A new 15-point "Safety Assessment," which includes categories such as human-machine interface, privacy and post-crash behavior, requires companies to submit technology before it is a deployed. That's a radical departure for the 46-year-old agency, which traditionally has set safety guidelines for automakers and dealt with failures via recalls.
Government officials have said that sharing such data not only with federal officials but also industry competitors is integral to providing consumers with the safest self-driving vehicles. Strickland says it remains to be seen how such an unprecedented collaboration will work out in practice.
"We have an open discussion and we'll find consensus," says Strickland, whose group members have been particularly aggressive with self-driving car research and testing.
"In the areas the department wants (data), it's worthy of discussion," he says. "That will, frankly, take a lot of conversations internally, and hopefully we'll find a way going forward. There is confidential business data that has to be respected, but safety is the number one priority."
The new policy document now enters a 60-day public comment period, during which those affected by the new directive can voice their concerns to its specifics. Once that period concludes, companies working in the autonomous vehicle sector will have four months to submit technological details to officials.
On a conference call with reporters Tuesday, senior government officials who could not be quoted by name said that the mission of the new policy was not to stifle or homogenize technological innovation, but rather to ideally create industry best practices that make consumers safe while preserving individual company breakthroughs.
Mike Nelson, a partner specializing in regulatory matters at New York law firm Sutherland Asbill & Brennan, is skeptical about aspects of the new policy.
"There are still several questions left unanswered," says Nelson, leading off with the government's ability to keep its promise to expedite automakers' request for legal interpretations and exemptions during the development process, which might not operate at a Silicon Valley pace. "The federal government is not known for its speed, as demonstrated by the fact that these guidelines were supposed to be issued this past summer."
His two other concerns involve the states' ability to quickly pass legislation related to self-driving cars and the thorny but critical issue of developing a legal framework for determining liability and damages in this area.
Government officials have made it clear that when a vehicle is operating solely under the guidance of software, that falls under the Department of Transportation's purview. But when a human is at the wheel, the states take over.
Such an approach, however, doesn't take into account more nuanced situations where a transfer of control between machine and human is instantaneous, as with driver assist functions such as Tesla's Autopilot. That system is currently in the spotlight as NHTSA investigates a May crash in Florida where Autopilot may have been in control as a Model S sedan hit a truck without slowing down.
The new policy makes clear that government officials have the right to "remove from the road" any vehicle whose technological systems might compromise driver and passenger safety. Tesla CEO Elon Musk recently announced an over-the-air update to Autopilot that is said to improve the partial self-driving system's ability to recognize objects in its path.
"A key theme (of the policy paper) is user awareness," says Aaron Steinfeld, associate professor at Carnegie Mellon University's School of Computer Science, which has educated many self-driving tech engineers. "There is guidance on informing drivers and riders about vehicle capability, clarity on how data will be used and consumer education and training."
For Coalition general counsel Strickland, the vested parties in the self-driving car revolution owe it to the 35,092 Americans — not to mention the 1.2 million globally — who died in car accidents last year to find common ground.
"We're excited to begin the process of working with the department and with states in order to make the deployment of this technology possible," he says. Overall, "the agency did a good job in reviewing the (autonomous car) ecosystem broadly."
Follow USA TODAY tech reporter Marco della Cava @marcodellacava