Driverless Cars

Benefits and Concerns

Proponents of autonomous vehicles insist that the technology will make the roads safer. A recent study by McKinsey & Co. showed that, in the United States alone, over 30,000 people die each year on the roads, and more than a million are injured. Over 90% of those incidents were caused by human error.

Another factor to consider as autonomous car technology continues to be introduced to the public are the incremental benefits that might be derived from its limited use. Perhaps, for example, automation will be introduced to only long-haul truckers while on interstates (much like the auto-pilot function on aircraft), allowing drivers to rest while goods are transported safely. Autonomous vehicles might also be initially restricted to ride-sharing situations, or serve to supplement public transportation in certain geographic areas. Of course, consideration can be given to elderly or disabled drivers who would also serve as “shot-gun” drivers in the vehicles (an option that certain technologies and government regulations may require in any event).

Indeed, at first glance, driverless cars seem to resolve many major concerns. As former General Motors chairman Bob Lutz pointed out: “The autonomous car doesn't drink, doesn't do drugs, doesn't text while driving, doesn't get road rage. Autonomous cars don't race other autonomous cars, and they don't go to sleep.” Instead, these vehicles utilize the stone-cold sober technologies of sonar, radar, image recognition, high quality laser sensors, assorted robotic features and exotic software and algorithms.

Sounds ideal. But there are many other things that autonomous vehicles don't do.

Autonomous vehicles don't yet know how to handle bad weather. Snow, rain, fog and other types of weather make driving difficult for driverless cars, which stay in their lanes and avoid collisions by using cameras and lasers that track lines on the pavement and their surroundings.

Autonomous vehicles don't know how to detect threats. Computer algorithms cannot control the behavior of pedestrians, bicyclists, or other drivers (who speed, pass illegally, etc.).

Autonomous vehicles also don't know how to weigh situations. They have no human ethical considerations, don't necessarily learn from experience, and don't know when it might be prudent to “break the rules.” For example, if a self-driving car's only options are to hit a wandering child or veer and strike a telephone pole, potentially injuring the car's occupants, what does it do? Should its computer give priority to the pedestrian or to the car's passengers? Can an algorithm determine whether to strike a vehicle to your left, or a stalled vehicle ahead (or brake hard and risk a rear end collision)? The list goes on …

'Safer Vehicles' Will Increase the Number of Liability Targets

Despite the outstanding concerns, a shift to driverless cars seems increasingly likely. And one thing that we know for certain is that there will be a plethora of new legal issues that arise from such a change in driving method. Decision-making is going to shift from a driver sitting behind the wheel, reacting impulsively, to teams of designers sitting behind computers, deliberately anticipating situations the vehicle may need to handle.

With a shift in decision-making comes a likely shift in liability. Courts will look to traditional negligence and product liability law to identify the appropriate remedies for any resulting injuries and damages. Plaintiffs in product liability lawsuits will ' as they have always done ' probably choose from among various “theories” of liability when seeking to recover damages.

Negligence

While car accidents at present can result from driver negligence, this will not necessarily apply in actions involving autonomous driving technology that is designed to replace the driver's actions and judgment. Plaintiffs in accident cases involving autonomous vehicles are more likely to claim that the accident was caused by a software or hardware defect in the autonomous system that did not properly evaluate emergency situations, road or weather conditions, or simply failed to prevent an accident under foreseeable conditions.

Perhaps a “shot-gun” driver may be blamed for not intervening at an appropriate time. Failure to keep a self-driving system “up to date” or to maintain it properly may implicate the owner of a vehicle. Of course, one must ask who will be responsible for the failure to maintain other component parts of the vehicle when they fail and contribute to the occurrence of accidents. (Indeed, to complete the circle, who will be responsible for the autonomous vehicle's failure to respond to emergency situations occasioned by component part failures?) Accordingly, actions will likely be brought not only against the driver, but also against the automobile's manufacturers, its component manufacturers, and its software designers. The list may be quite extensive when considering that negligence standards look to decipher what is reasonable and/or foreseeable under various new situations that may arise.

Defective Design

As suggested previously, numerous additional design issues will likely arise involving software, hardware and the anticipated involvement (or lack thereof) of a human driver. The new structure of vehicle design in the age of self-driving should easily fit into an analysis of whether “the product” was reasonably safe when used in the intended manner, as well as a “risk-utility analysis” which, among other things, will determine if the design's utilities outweigh its risks or whether there are readily available alternative designs. At the moment, while the technology is in its relative infancy, it remains to be seen which design parameters will become “state of the art.”

And what about consumer expectations; does a plaintiff need to claim more than “I expected the vehicle to drive me safely from Point A to Point B”? Possibly not. Does this imply that manufacturers may be headed toward a status akin to an insurer for the remaining types of lawsuits that may arise in the future (especially when there is no driver to point the finger at after an accident)? To be sure, the insurance industry is bracing for a major change. The advent of fully autonomous vehicles will dramatically change the insurance industry. Big-Four auditor KPMG's October 2015 research report on the effects of autonomous vehicles estimates that they could reduce the number of accidents by 80% by 2040, dramatically reducing the frequency of insurance claims, and result in a percentage increase of the remaining claims to manufacturers or designers.

The reduction in accidents will also result in lower premium rates and, therefore, less profit for the insurance companies. The real question, however, becomes this: Do individual autonomous vehicle owners need personal auto insurance when there is no “driver” to insure?

Manufacturing Defects

Manufacturing errors will likely be a continuous source of claims. Mistakes on the production line will not disappear. In terms of self-driving technology, even when a design is sound, manufacturers can be liable for defects. For example, if a manufacturer accidentally ships a vehicle with an older version of the software containing a flaw not present in the newer version, or if errors traditionally associated with software malfunctioning occur, leading to accidents, a manufacturer could be held strictly liable for injuries sustained by an individual.

Failure to Warn; Failure to Update

Of course, there are a number of additional liability risks that manufacturers will need to address with driverless cars. Manufacturers that fail to provide adequate information regarding the risks of using a product can be liable for failure to warn when an injury attributable to this lack of information occurs. Issues will arise when circumstances require driver interaction. Was the driver/user properly informed as to how to use the technology, or advised concerning the inherent risks involved?

Moreover, manufacturers that become aware of software problems will need to act quickly to provide upgrades, while at the same time ensuring that the upgrades are appropriately tested before they are released. Allowing a manufacturer access to cars that are already on the road to provide software updates raises the issue of cyber-security in autonomous vehicles. When cars are being run by computers, there is the inherent risk of “hacking.” Already, several incidents of vehicle hacking have occurred, including a report that a Tesla was hacked by researchers, and a report that engineers hacked a Toyota Prius and Ford Escape to take control of essential vehicle functions. Allowing autonomous vehicles on the road with the potential to be hacked is not only dangerous to the consumer, but could lead to numerous legal issues for the manufacturer.

Warranty/UCC

Manufacturers will continue to be governed by the Uniform Commercial Code (UCC) when selling their products. Express warranties relating to autonomous vehicles will likely be issued to the buyer at the point of sale, and they will come about through advertising the function of the technology, promoting new technology at the expense of older technology. Such warranties will ultimately be created by a manufacturer's need to differentiate its product in a competitive market. Additionally, the implied warranties of merchantability and fitness for a particular purpose will also apply to technology relating to autonomous vehicles.

Government Guidelines and Regulations

As early as 2013, the federal government entered the discussion when the National Highway Traffic Safety Administration (NHTSA) issued a Policy Statement defining what it characterized as “levels” of automation in vehicles. Since then, NHTSA has recently begun the process of conducting public hearings with the goal of developing new guidelines for self-driving vehicles by the end of this summer. As expected, the automotive industry has gone public with its concerns that NHTSA is moving too quickly, while consumer groups are pushing for more guidance at the federal and state levels. California has already proposed regulations which would require that self-driving cars be equipped with steering wheels and brake pedals. Mark Rosekind of NHTSA summed up regulators' concerns when he commented that people “just keep putting stuff out on the road with no guidance on how do we do this the right way.”

Other Concerns

Regulators aside, not everyone is a fan of the idea of driverless cars. For one thing, autonomous vehicles could destroy jobs: “Nearly 3 percent of the U.S. workforce ' about 3.6 million people ' feed their families by driving taxis, ambulances, buses, delivery trucks, tractor-trailers and other vehicles.” Steven D. Levitt and Stephen J. Dubner: Think Like a Freak (May 2014, William Morrow and Company). Likewise, there is concern that if commercial trucking is completely automated, it would be economically devastating for small towns in America that thrive from supporting the long-haul trucking industry.

Conclusion

Though the complete transition to driverless cars may take years, it is clear that automobile manufacturers and leading technology companies are committed to the success of autonomous vehicles. As the consumer transitions to autonomous vehicles, so must the legal world. While autonomous vehicles will hopefully reduce the amount of accidents, those that remain will be governed by products liability standards ' shifting the blame from the driver to the manufacturer. Understanding how theories of products liability can be applied will be essential to best advise your manufacturing clients on how to proceed as we move forward into this new era.

Alan D. Kaplan is a Partner and Chair of the Product Liability Group at Herrick, Feinstein LLP in New York City. Robert Sanzillo is an associate in the litigation department at the firm.