COMPENSATION AT THE CROSSROADS: AUTONOMOUS VEHICLES & ALTERNATIVE VICTIM COMPENSATION SCHEMES.

• Author: Pearl, Tracy Hresko

TABLE OF CONTENTS

INTRODUCTION 1830 I. BACKGROUND 1835 A. Levels of Automation 1836 B. Consumer Adoption 1839 C. Advantages of Autonomous Vehicles 1841 D. Risks Associated with Autonomous Vehicles 1845 II. LIABILITY AND COMPENSATION MODELS 1849 A. The Tort System 1851 1. Advantages of the Tort System 1852 2. Drawbacks of the Tort System 1853 B. Victim Compensation Funds 1857 1. Advantages of Victim Compensation Funds 1858 2. Disadvantages of Victim Compensation Funds 1861 C. Finding a Venue for Autonomous Vehicle Crash Cases 1863 III. CATEGORIZING VICTIM COMPENSATION FUNDS 1866 A. Quasi-Judicial Funds 1867 B. Non-Judicial Funds 1870 1. Public Funds 1870 2. Private Funds 1872 3. Charitable Funds 1874 IV. A FUND FOR AUTONOMOUS VEHICLE CAR CRASH VICTIMS 1876 A. Proposed Coverage Limitation 1878 B. Proposed Source of Funding 1879 C. Proposed Administrator 1882 D. Proposed Participation Requirements 1883 1. Victim Participation Requirements 1883 2. Vehicle Manufacturer Participation Requirements 1885 E. The Private Insurance Overlay 1888 CONCLUSION 1889 "America is at a historic turning point for automotive travel. Motor vehicles and drivers' relationships with them are likely to change significantly in the next ten to twenty years, perhaps more than they have changed in the last one hundred years." --National Highway Traffic Safety Administration, 2013 (1) INTRODUCTION

On February 14, 2016, a white Lexus SUV drove down El Camino Real in Mountain View, California. (2) After signaling that it wished to turn right, the Lexus moved into the far right lane just before the intersection of El Camino Real and Castro Street. (3) However, sandbags situated around a storm drain blocked the car's path and forced it to stop. (4) The vehicles in the other lanes were stopped at a red light, so the Lexus had to wait until the light changed and the traffic flow resumed before attempting to inch out around the sandbags and into the left lane. (5)

At first glance, this driving scenario falls far short of being exciting, novel, or even particularly interesting. Most drivers likely encounter similar situations on a weekly or perhaps even daily basis, and handle them adeptly without much thought or anxiety. Rogue garbage cans that must be avoided on residential streets, construction equipment or barriers that block portions of highways, and cars parallel parked too far from the curb are all part of the American Driver's day-to-day landscape. Yet, the sandbags and the Lexus situation was profoundly different. It was a sea change in that landscape. The Lexus was driving itself. (6)

Like a human driver, the Lexus had to decide when it was safe to move the vehicle into the left lane to travel around the sandbags. (7) Its algorithms dictated that the vehicle wait for several nearby vehicles to pass before making its attempt. (8) Unfortunately, the car's algorithms misjudged the traffic flow: as the vehicle reentered the left lane, it sideswiped a public transit bus. (9) Fortunately, no humans were injured. (10) Both vehicles were traveling slowly at the moment of impact, so the consequences were fairly nominal. (11) The Lexus sustained damage to its front fender, left front wheel, and driver's side sensors. (12) The bus escaped with even less damage. (13)

Google, the designer of the autonomous Lexus, quickly claimed "some" responsibility. (14) "[I]f our car hadn't moved there wouldn't have been a collision," its monthly report stated. (15) Google also seemed to believe, however, that the vehicle's mistake had not been particularly egregious. The (human) Google employee who was monitoring--but not controlling--the Lexus at the moment of the crash noted that he had seen "the bus approaching in the left side mirror but believed the bus would stop or slow to allow the Google [autonomous vehicle] to continue." (16) If the human monitor also misjudged the situation, perhaps the autonomous vehicle's failure was not particularly troublesome.

Regardless of whether poor programming or merely a minor and unavoidable blip on Google's otherwise impressive safety record caused this fender bender, (17) this incident highlights the arrival of a new and profoundly novel legal issue: who should be liable (if anyone) and how should victims be compensated (if at all) when autonomous vehicles cause injury? Although the Lexus and the bus case did not result in litigation, we should expect such cases to arise and to do so at any moment. Semi-autonomous (partially driverless) cars are already available to consumers and on U.S. roads, (18) and fully autonomous ones continue to be tested on public roads in preparation for arrival on the consumer market within the next five years (if not significantly sooner). (19) As many legal scholars have wondered: is the American legal system ready? (20)

The answer to this question has implications far beyond the resolution of individual autonomous vehicle crash cases. Whether the American legal system can handle these cases fairly and efficiently implicates (1) the likelihood that consumers will adopt this new technology, and (2) the rate at which they will (or will not) do so. These implications should concern law and policy makers immensely. If autonomous cars stand to drastically reduce the number of fatalities and injuries on U.S. roadways--and virtually every scholar believes that they will (21)--then failing to establish a functional adjudication and compensation process risks stymieing adoption of this technology and leaving more Americans at risk of dying at the hands of human drivers.

The problem, of course, is that autonomous vehicles pose "a plethora of new and unique legal issues, which will need to be analyzed to facilitate the adequate transition of this new technology to the marketplace." (22) Chief among these are the legal implications of automation itself. Given that, as one scholar has said, "[t]he entire history of human laws has assumed that people make decisions," (23) handing those decisions over to an algorithm places lawyers and judges into a situation in which "[w]e currently have no legal framework for... liability." (24) Worse yet, the development of automated vehicles is already far ahead of the development of the law in this area, and automated vehicle technology continues to advance at rates that can often seem exponential. (25) We are thus in a situation in which we need to develop jurisprudence in and around a technology that challenges many of American jurisprudence's most fundamental assumptions. We must do so, moreover, extraordinarily quickly or risk hampering innovation and slowing adoption of a technology that will likely save tens of thousands of lives each year in the United States alone. (26) In short, the stakes are extremely high and the time extremely limited.

In this Article, I explore the liability issues posed by accidents involving autonomous vehicles and propose a way in which we can both compensate injured victims while also creating time and space for the civil justice system to develop a robust jurisprudence in and around the use and development of these vehicles. It is my belief that funneling autonomous vehicle crash cases into a specially designed, no-fault, quasi-judicial victim compensation fund is a sensible way to do so. Such a fund could both protect autonomous car designers and manufacturers from high levels of uncertainty about their exposure to liability and assure consumers that they will be compensated fairly and quickly if an autonomous vehicle harms them. (27) Although it is not my intention to wholly replace the tort system in cases involving autonomous technologies (and certainly not my intention for this Article to serve as a referendum on the value of the tort system as a whole), my hope is that my proposed stop-gap solution will give courts "breathing room" to adapt products liability law to the brave new world of automation and artificial intelligence while simultaneously offering victims and manufacturers greater confidence that autonomous vehicle crash cases can be handled fairly and efficiently if and when they occur. (28)

The National Vaccine Injury Compensation Program (NVTCP) provides an excellent example of the type of quasi-judicial compensation fund that could be well-suited for autonomous vehicle crash cases. (29) Although victim compensation funds in the United States have taken various forms--ranging from the federally funded and publicly administered September 11th Victim Compensation Fund (30) to the privately funded and privately administered Deepwater Horizon Oil Spill Trust (31)--the NVICP's quasi-judicial setup, non-adversarial process, and reliable funding mechanism make it a better model for adjudicating autonomous vehicle crash cases. (32) Additionally, as discussed at length in this Article, although the NVICP has its shortcomings, it has a far better and more extensive track record than many of the other types of victim compensation funds created and administered over the last several decades in the United States, and thus appears more likely to succeed in a context such as this one. (33)

In Part I of this Article, I describe the development of autonomous vehicles in the United States, the way in which the federal government has chosen to categorize them, and the benefits and advantages they offer to consumers over human-driven vehicles. In Part II, I explore and analyze the two courses that the United States could take in handling autonomous vehicle crash cases: (1) sending them through the existing civil justice system and applying existing products liability and negligence jurisprudence, or (2) creating a victim compensation fund to handle them. In Part III, I explore the numerous victim compensation fund design options: quasijudicial, public, private, and charitable. In Part IV, I propose a model victim compensation fund for injuries arising from the use of autonomous vehicles based on the NVICP, propose a...