The Fast Lane: What Driverless Cars Mean For Innovation And Risk
It’s coming quickly down the road: a world where we can get in a car anytime we want to but don’t own one. Where we’re all passengers. Where accidents are drastically reduced, and we don’t have to worry about dangerous drivers on the road. Where we can join conference calls or draft a report on our morning commute.
Cars are changing more quickly and drastically now than at any other point in their history. And it’s no wonder: from 2014 to 2017, startups, automakers and other stakeholders invested an estimated $80 billion into autonomous vehicle (AV) technology. Still, it’s unclear when this technology will achieve widespread consumer adoption.
Alongside continued investment, three primary trends are shaping the future of cars: electric technology, autonomy and mobility services. The convergence of these three trends promises to completely revolutionize how consumers and companies think about cars as they relate to insurance, risk, and safety.
Automakers and entire industries – from transportation and logistics to insurance – will be forced to adapt to the changing landscape. How did we get here? And what is the route forward?
As we consider the implications of the evolution of the automobile, it’s helpful to understand the context for this change and imagine the future possibilities.
The History And Evolution Of Cars
Cars have been constantly evolving since the first Model T rolled off the production line in 1908. But the current phase of innovation has created some confusion among the general population. Jillian Slyfield, Digital Economy Practice Leader, Risk Consulting and Insurance Services at Aon, said that when it comes to cars, “People either picture flying cars from ‘The Jetsons’ or the historic Model T. What’s not as clear is the interim and the various risks as cars move toward autonomy.”
Many changes to cars over time have had the dual effect of improving driver experience and forcing the industry and policymakers to balance innovation and safety. For example, power steering, which became a common feature for drivers in 1956, made maneuvering a car easier and potentially improved safety. The addition of cruise control, in 1958, originally called Speedostat, both improved user experience and paved the way for today’s autonomous features. By 1984, you could expect your car to come equipped with air bags. While they overall improve the safety of vehicles, they also come with concerns such as child safety.
Looking ahead, observers believe that widespread adoption of fully autonomous (Level 5) vehicles is at least a decade away due to barriers to adoption, such as infrastructure development, legislation and consumer acceptance. Strides are being made, however: vehicles with Levels 4 (high automation) and 5 functionality are already being tested on the roads, and legislation specific to AV has passed in 21 states. Further, many automakers have announced the expected rollout of cars with some level of automation (mainly 3 and 4) for commercial purchase over the next few years. Volvo, for example, is targeting a release date in 2021 for a Level 4 vehicle. Some industry experts predict up to 21 million autonomous cars could be sold globally in 2035 and that more than half of U.S. traffic could be autonomous by 2050. Regardless of the precise timeline, it is clear that autonomous-vehicle technology will continue to improve and it’s only a matter of when, not if, it becomes ubiquitous.
The Road To Autonomy
What will full autonomy look like? Slyfield explains that this process will have many steps. “We’re not looking at an overnight change to driverless cars. Human drivers will likely intermix with autonomous vehicles on the road, and there will be various stages toward full autonomy.”
Mike Stankard, Automotive Practice Leader, Aon, points out, “We’re experiencing highly autonomous vehicles today.” And as time goes on, more and more pieces will become automated.
To understand this point, it’s helpful to think of autonomous capabilities along five levels. For example, Tesla Autopilot, available in a number of Tesla vehicles, and Volvo Pilot Assist, available in the XC60, currently feature Level 2 or partial automation, meaning they can steer, accelerate and brake under some circumstances. The Audi Traffic Jam Pilot technology featured in Audi’s 2019 A8 sedan (available in Europe at the end of this year), achieves Level 3 or conditional automation, in that it can manage almost all driving functions, including monitoring the environment. Drivers using this technology just need to be available to take over if the car encounters a scenario it doesn’t know how to handle.
Data And Analytics: Driverless Cars And Improving Safety
Much of the excitement around autonomous vehicles is their potential to dramatically reduce fatalities on the road. The National Highway Traffic Safety Administration (NHTSA) reports that, of the approximately 35,000 annual auto deaths, 94 percent are tied to human choice or error. In fact, according to the U.S. Department of Transportation, in 2015, 94 percent of the year’s estimated 2.2 million crashes were the result of driver error while just 2 percent were the fault of the environment, another 2 percent were the fault of vehicles, and the remaining 2 percent were attributed to unknown critical reasons.
As cars continue to become more automated, the percentage of crashes through driver fault will shrink drastically. “Today if there’s a car accident,” Slyfield notes, “much of the liability falls on the driver and some on the product. But if you fast-forward to the future, more of the liability or fault falls on the technology and manufacturer of the vehicle.”
And it’s not just about liability. As the number of autonomous vehicles increases, “the number of accidents anticipated goes down substantially,” Slyfield notes.
While full adoption of Level 5 vehicles is potentially some way off, experts concede that onboard computers will increasingly take over, particularly for city driving. With these computers, cars will become more and more proficient at communicating with one another and control centers, harnessing data to improve safety and traffic flow.
By broadcasting their position, speed and steering-wheel position to nearby cars through wireless technology, for example, other cars are better able to understand their surroundings, identify any impending perils, and determine which corrective actions to take. And through communication with control centers – for instance, cars broadcast their location and receive the locations of other vehicles – cars can choose routes with less traffic and congestion, not only speeding up travel times but also decreasing the risk of accidents. But not just road users benefit from improved safety. The application of ADAS (advanced driver-assistance systems) on a large scale would also prevent a lot of damage for insurers and employers. The latter are now seeing their mobility costs rise each year because of higher car insurance premiums and the significant costs due to employee absenteeism as a result of serious traffic accidents.
The cause of the remaining traffic accidents will be much easier to diagnose. According to Stankard, the black-box technology of these vehicles “will bring near more certainty in determining each vehicle’s contribution to a crash event, which will can considerably accelerate claim resolution and reduce unnecessary legal costs.”
Increased Data And Impact On Insurance Premiums
But significant ramifications exist for car insurance car liability. “Thanks to the countless data that the connected car collects, the amount of knowledge and understanding from the frequent use of ADAS will grow rapidly,” says Evert-jeen van der Meer, Industry Director, Aon Risk Solutions. “It will make driving cars driving even safer and increase confidence in fully autonomous cars. Fewer traffic accidents will also reduce insurance claims, which will increase the insurability of highly autonomous cars again. Unlike many global mobility providers, insurers do not realize a healthy return on car insurance.”
Stankard points out that “46 percent of all premiums . . . drive back to auto and motor insurance.” If the 94 percent of accidents related to driver error will be drastically reduced with the advancement of autonomous technology, the percentage of these premiums will shrink significantly. The industry must consider, then, how to replace those premiums and what new needs consumers will have in the place of standard auto insurance.
A Shifting Risk Landscape
“There’s nothing definitive around when this technology will become widespread,” Slyfield says, “but as this cultural adoption happens, we have to ask ourselves how to address new risk issues and processes with clients and consumers.” One of these new issues? Ownership in the on-demand economy. Going forward, the private sector and policy makers will have to seriously consider the implications of fewer and fewer people owning vehicles.
Some traditional auto manufacturers are already making moves to position themselves for the changing landscape. For example Toyota, eager to explore new mobility services and strategies in Asia, has invested $1 billionin the ride-hailing app Grab so it can monitor traffic and driving habits. GM made a similar move in 2016 when it launched its own car-sharing service, Maven, which it developed with a team of connected-car technology and ride-sharing professionals. And Ford’s Smart Mobility segment has made significant investments in self-driving cars and multimodal transportation, among other areas.
Although it remains unclear when self-driving cars will achieve mass adoption, risk will likely shift dramatically.