Article: Legal Issues Raised by the Driverless Vehicle Revolution – PART 2

Part 1 of this article, published in the December 2015 Business Litigation Report, described emerging driverless vehicle technology and related patent and patent ligation issues. Part 2 continues this discussion with the potential impact of the driverless vehicle revolution on products liability law and data security.

Despite the potential of driverless and driver-assisted vehicles to dramatically reduce the frequency and severity of accidents, there is cause for concern that these vehicles may “increase the liability exposure of vehicle manufacturers.” Gary E. Marchant & Rachel A. Lindor, The Coming Collision Between Autonomous Vehicles and the Liability System, 52 Santa Clara L. Rev. 1321, 1149 (2012) (emphasis in original). As automation increasingly supplants the driver’s role in avoiding and mitigating accidents, liability for accidents can be expected to shift from drivers to automobile manufacturers. Thus, over the next 25 years, personal automobile insurance lines may shrink while commercial and products liability lines expand – perhaps dramatically. KPMG, Keeping Up with the Pace of Change: Demands by Customers Are Driving the Property & Casualty Agenda, at 20, available at http:// bit.ly/1R5ojsj.

The driverless vehicle revolution could disrupt the generally accepted assumption that accidents are “a frequent and inevitable contingency of normal automobile use.” Larsen v. Gen. Motors Corp., 391 F.2d 495, 502 (8th Cir. 1968).  Indeed, Toyota described the “ultimate goal” of its driverless car program as creating “a car that cannot be responsible for a collision.” Richard Waters, CES 2016: Toyota poaches Google exec to help lead AI effort, Fin. Times, Jan. 5, 2016, at 15. As automobile accidents become less frequent and are no longer perceived as inevitable, their mere occurrence could be perceived as a strong indication of a product defect.  Historically, such paradigm shifts have resulted in increased liability exposure. For example, the commercial availability of automobiles gave rise to modern products liability law and improvements in automobile safety spawned the “crashworthiness” doctrine.  This article recalls these significant landmarks from the history of automobile products liability law, explores the current legal landscape, and attempts—despite low visibility conditions—to survey the driverless road ahead.

Two Landmarks in the History of Automobile Products Liability Law.
Modern automobile products liability law began a century ago with MacPherson v. Buick Motor Co., 217 N.Y. 382 (1916). In MacPherson, the plaintiff sued an automobile manufacturer for injuries sustained in an accident caused by a latent defect in his car’s steering wheel. Although a third party had supplied the steering wheel, the manufacturer could have discovered the defect through a reasonable inspection prior to sale. Before MacPherson, the plaintiff would have been expected to sue the dealer—with whom he was in privity of contract—rather than the manufacturer. And even against the dealer, recovery would not have been assured because products liability was generally limited to “inherently dangerous” products like poisons or explosives. Yet Judge Cardozo upheld judgment for the plaintiff in an opinion that, in one fell swoop, eliminated the privity requirement and deemed automobiles to be ordinary products that become “inherently dangerous” when negligently designed or manufactured. MacPherson thus established the central principles of modern automobile products liability doctrine and triggered what some commentators called a “liability explosion.” Lawrence M. Friedman, Law in America: A Short History (2002), at 129-30. 

More than 50 years later, the “crashworthiness” doctrine marked another expansion of automobile products liability law. In 1968, the Eighth Circuit held in Larsen v. Gen. Motors Corp. that an auto manufacturer has “a duty to use reasonable care in the design of its vehicle to avoid subjecting the user to an unreasonable risk of injury in the event of a collision.” 391 F.2d at 502. This doctrine applies regardless of what caused the collision and allows recovery for “enhanced injuries”—i.e., injuries over and above what would have occurred had the vehicle been crashworthy—even if the collision itself was the plaintiff’s fault.

In the wake of Larsen, a majority of states explicitly adopted the crashworthiness doctrine.  See  Larry E. Coben, Crashworthiness Litigation § 1:2 (2d ed.). Likewise, the implied warranty of merchantability has since “been held to embrace some measure of ‘crashworthiness’,” as first recognized in Larsen. 18 Williston on Contracts § 52:112 (4th ed.).

The MacPherson and Larsen courts thus expanded products liability law in response to evolving consumer expectations and emerging automotive technologies. Whether courts will respond similarly to autonomous vehicle technology, however, is uncertain. To date, courts have consistently recognized the general principle stated in Larsen that “an automobile manufacturer is under no duty to design an accident- proof or fool-proof vehicle.” 391 F.2d at 502.

But just as MacPherson and Larsen fundamentally altered the products liability landscape, this general principle too could change, possibly ushering in a new era of automobile products liability. Indeed, our acceptance of auto accidents as inevitable may one day be viewed “as a historical aberration, present only during the century or so when technology enabled the mass production of cars, but not of highly automated systems to help drive them safely and reliably.” John Villasenor, Brookings Institution, Products Liability and Driverless Cars: Issues and Guiding Principles for Legislation, at 3 (April 24, 2014). Ironically, a sharp decline in auto accidents could provide fertile ground for the next landmark products liability decision.

The Current Landscape of Safety Regulation and Products Liability.
The laws and regulations governing the safety of autonomous vehicles are, unsurprisingly, in their infancy. While the federal government has the authority to regulate the design and operation of vehicles used on public roadways, it has yet to formally regulate autonomous vehicle technology.  Administrative agencies, however, have signaled their interest in this area. For example, in 2013 the NHTSA first outlined its plans to make recommendations as to the testing, licensing, and regulation of autonomous vehicles. And in December 2015, the NHTSA announced that it plans to revise its 5-star safety rating to take into account whether vehicles are equipped with active crash-avoidance technologies and to evaluate the performance of those technologies. These revised ratings should begin to appear with 2019 model year vehicles.

A majority of states have considered legislation governing driverless cars, and a handful—including California, Florida, Michigan, Nevada, Tennessee, and the District of Columbia—have already passed legislation. The state bills generally track legislation first enacted in Nevada and share several common features, including definitions of what qualifies as an autonomous vehicle and who is considered its “driver.” These bills also typically authorize testing of autonomous vehicles on state roads, while requiring a driver capable of taking over manual operation of the vehicle, certain minimum insurance coverage, and the ability to comply with state traffic laws.

Many of the passed and proposed state statutes contain safe harbor provisions under which manufacturers of vehicles converted to autonomy by a third party are not liable for an injury that results from that conversion unless the defect that caused the injury was present in the vehicle as originally manufactured. However, these safe harbor provisions typically do not protect commercial sellers or distributors, whose status in the stream of commerce has traditionally made them subject to various theories of liability for damages caused by product defects, including strict products liability. Standard defenses, such as product modification, product misuse, or lack of causation, remain available to sellers or distributors.

The Road Ahead: Products Liability Issues in the Coming Age of Driverless Vehicle.
Traditional Products Liability Principles.  In most instances of a driverless vehicle crash, personal injury claimants will likely rely on traditional legal theories. Garden variety automotive accident claims almost universally sound in negligence. But the negligence framework, which assesses the reasonableness of the driver’s actions, is ill-suited for claims involving autonomous driving technology, which is designed to supplant the driver’s actions and judgment. Consequently, plaintiffs in accident cases involving autonomous vehicles are more likely to claim (and may have no choice but to claim) that the accident was caused by a software or hardware defect in the autonomous system rather than by driver error. Thus, accident cases that would have previously been brought only against the driver might also give rise to—or be replaced by—products liability claims against automobile manufacturers.

There is always some uncertainty as to how existing legal frameworks can be applied to emerging technologies. When presented with novel technologies, courts often try to make analogies to legal constructs governing existing technologies. Here, however, cases involving the most analogous existing technologies— autopilot systems for airplanes and navigation systems for long-distance maritime vessels—apply standard products liability legal constructs with little variation. Accordingly, at least early on, we can expect courts to apply the familiar principles of products liability law to address issues raised by driverless vehicles.

Who Will Be Held Liable.  The unanswered question of who will be held liable for autonomous driving accidents is perceived by many as a potential roadblock to the development and eventual widespread adoption of the technology. Some early innovators in driverless technology have decided to try to address this concern proactively: for instance, Google, Mercedes Benz, and Volvo have all stated that they will accept responsibility for any accident that is caused by a flaw in their autonomous driving technology. While such statements do not prevent plaintiffs from suing drivers or vehicle owners, they may provide some comfort to consumers and thus encourage the purchase and use of driverless technology.

Despite these proactive measures by certain automobile manufacturers and technology companies, there can be little doubt that “[a]utonomous vehicles will complicate the already complicated entanglements between insurance providers, plaintiffs, drivers/owners named as defendants, and manufacturers.”  John Villasenor, Brookings Institution, Products Liability and Driverless Cars: Issues and Guiding Principles for Legislation, at 15 (April 24, 2014). To be sure, though, until full autonomy is achieved, auto accident litigation involving autonomous driving where the driver still plays a role will present “complex questions of liability shared by both the human driver and autonomous vehicle technology providers.” Id. at 15.

Tesla, the Northern California-based electric car company, has suggested a toe-in-the-water approach to offering automated driving technologies while making a human driver responsible for the safe exercise of that technology: drivers will have to activate their turn signal to trigger their vehicle’s autonomous passing function.  The Tesla proposal contemplates making drivers responsible for the timing of passing maneuvers, while the execution of those maneuvers would be fully automated. In theory, by choosing when to engage the turn signal, the driver “acknowledges road conditions are appropriate for a passing maneuver and … takes responsibility for the consequences.” Mike Ramsey, Who’s Responsible when a Driverless Car Crashes? Tesla’s Got an Idea, Wall St. J., May 13, 2015.

It should be noted that even manifest driver error in an accident may not fully exculpate the autonomous technology provider.  For example, some crash- avoidance technologies—such as  lane departure warning and blind-spot detection—are designed not to substitute for the driver, but rather to monitor driver behavior, detect driver inattention, and alert the driver when necessary. Depending on the nature of the accident, it may be possible to characterize some instances of driver error as a failure of those technologies to operate as they should, thus reducing—or perhaps even eliminating—driver liability. Savvy lawyers will no doubt make such arguments in defense of their clients.

Given the uncertainty over the issue of to whom to assign liability for autonomous vehicle-involved accidents, it should come as no surprise that no states have rescinded their requirements that, to date, owners of automotive vehicles carry at least certain amounts of liability insurance. In fact, many of the state statutes that have been enacted to govern autonomous vehicle technology not only require drivers to carry insurance, but also specify the minimum amounts of insurance that must be maintained to use driverless vehicles on state roads.

“No Injury” Products Liability Claims and Class Actions. Suppose a driverless or driver-assisted vehicle is involved in an accident that arguably should have been prevented by the vehicle’s autonomous safety features. Could owners of the same vehicle model invoke that accident as proof of a design defect and sue for economic damages even though the defect had not manifested in their own cars? The majority rule of products liability rejects such “no injury” claims. In one leading case, the plaintiffs alleged that their anti-lock braking systems were defectively designed. Although their brakes had not actually failed, plaintiffs sought economic damages for overpayment and lost resale value. The Eighth Circuit affirmed dismissal, holding that “[w]here … a product performs satisfactorily and never exhibits an alleged defect, no cause of action lies.” Briehl v. Gen. Motors Corp., 172 F.3d 623, 628 (8th Cir. 1999). However, a minority of courts in other circuits have recognized such claims and even allowed them to proceed as class actions. See, e.g., Chamberlan v. Ford Motor Co., 369 F. Supp. 2d 1138, 1147 (N.D. Cal. 2005).

Data Security and Breach. To perform many of their functions, driverless cars will need to communicate wirelessly  with vehicle manufacturers and technology providers, and ultimately with other vehicles and roadside resources. The Department of Transportation is currently involved in developing standards for vehicle-to-vehicle communications and the FCC has dedicated a specific bandwidth range to enable vehicles to communicate wirelessly with one another and with roadside infrastructure.

As more automobiles communicate wirelessly, the risk that these systems will be “hacked” increases. Already, several incidents of vehicle hacking have occurred, including a 2015 report that a Tesla Model S was hacked by researchers, and a 2014 report that two DARPA engineers hacked a Toyota Prius and Ford Escape to take control of essential vehicle functions. While these cases required physical access to the vehicle’s systems, the potential for a malicious hack has already led to at least one lawsuit against major automakers alleging that “hackable” vehicles are legally defective. Although such allegations are subject to challenge for failure to plead actual harm, they leave little doubt that future lawsuits will include allegations that vehicle manufacturers are to blame for accidents or exposure of personal data that result from hacking