By
Peter Hart

May 19, 2017, 11:19 AM EDT

Law360, New York (May 19, 2017, 11:19 AM EDT) —
Peter HartFor companies with drivers on the payroll, the shift to fully autonomous vehicles could usher in massive reductions in potential liability. After all, driver error is the leading cause of accidents on U.S. roadways.

Simply put, bad decisions by human drivers are to blame for everything from innocuous fender benders to nightmarish fatal collisions. (Last year, an estimated 40,200 people died in traffic accidents across the United States; many observers cited a quintessentially human misstep — distracted driving — as a significant factor.)
When perfected forms of autonomous driving technology replace error-prone human drivers, many are betting that businesses such as trucking firms, delivery services and shuttle operators will face dramatically fewer legal settlements and court battles triggered by vehicular accidents. The stated goal of the most ardent advocates of the technology is to one day cut traffic fatalities to zero. However, it will take years, if not decades, for driverless cars and trucks to live up to their promise of making our roads collision-free.

It would be difficult at best to predict precisely how autonomous vehicles will affect liability and litigation, but current trends provide some clues. Take the increasing sophistication of the specialized Electronic Control Modules (ECMs) used in commercial trucking. These devices are the trucking industry’s version of the “black box” data-collectors that play such key roles in the reconstruction of aviation accidents.

Located in the cab of the truck, ECMs track the vehicle’s speed, braking data, gas consumption and more. Prior to the adoption of ECMs in the 1990s, efforts to reconstruct accidents relied primarily on eyewitness reports, skid marks and the like.

Today, thanks to the wealth of post-event information that is now available from ECMs, it is easier to scientifically reconstruct traffic accidents. The devices are getting more sophisticated all the time: Next-generation ECMs might even communicate vehicles’ speeds to dispatchers in real time, allowing them to call lead-footed drivers and order them to slow down.

A Boon to Defense Teams

Compared to the rich data that will be available from driverless cars after an accident, ECMs are relatively primitive. Autonomous vehicles are loaded with advanced sensors such as beacon-based locational systems, stereo “vision,” lidar, GPS and inertia-measurement devices.

Data from these systems will provide unprecedented detail and accuracy in accident-reconstruction. The rich datasets recorded by these cars will prove invaluable to defense teams in suits filed against manufacturers or owners/operators of autonomous vehicles such as trucking firms, ecommerce delivery services and Uber-style transportation companies, to name a few.

The envisioned future in which all vehicles on the roads are controlled by computers is almost certainly decades away. The interim period will be one in which mistake-prone and increasingly distracted humans share the roadways with fleets of autonomous cars and trucks.

So far, at least, preliminary data from the likes of Google (other companies road-testing the technology include Uber, Otto, Tesla, Volvo, Toyota and BMW) does seem to indicate that even the early “1.0” iterations of driverless vehicles are safe relative to the error rates of human drivers.

In 2016 alone, the Google autonomous vehicle project Waymo logged 635,000 miles on the public roads of California. Backup drivers in these vehicles reportedly needed to take control of the cars only 124 times during that period, or about once every 5,000 miles.

“We’ve self-driven more than 2 million miles mostly on city streets,” Waymo states on its site. “That’s the equivalent of over 300 years of human driving experience, considering the hours we’ve spent on the road. This builds on the 1 billion miles we’ve driven in simulation in just 2016.”

Most of the accidents thus far have been the result of humans crashing into driverless cars or otherwise causing accidents with them, not the other way around. The first exception reportedly occurred in 2016, when a Google car pulled in front of a bus to avoid some sandbags, causing a fender bender.

In another high-profile accident — the first fatality involving a semi-autonomous vehicle, in this case a Tesla Model-S — regulators found no fault with the functioning of Tesla’s Autopilot system. The driver, they said, had taken his eyes off the road for an extended period of time and was using the system improperly. More recently, the fault in a crash involving an autonomous Uber reportedly lay with the human driver of the other vehicle.

Shifting the Burden

Undoubtedly, more collisions will occur in coming years as more autonomous and semi-autonomous vehicles mingle on U.S. roads with cars and trucks driven by people. If predictions of dramatically improved safety do hold, then humans, and not driverless vehicles, will typically be at fault in these accidents. This will bring tremendous liability relief to companies that choose to eschew drivers and employ the technology.

This is not to suggest, however, that all liability risk will vanish. While the majority of accident-related litigation today stems from driver error, many lawsuits (including class actions) target manufacturers for mechanical failures such as faulty steering wheels, braking systems or accelerators.

Another possibility is that, at least initially, autonomous vehicles could cause a rash of accidents as manufacturers strive to discover and eliminate the remaining bugs in these systems. It is easy to imagine the kinds of lawsuits that could occur: “The manufacturer claimed the car would be safe in the snow,” the plaintiff tells the jury, “but the car was totally confused in that blizzard and caused a horrific crash.”

Insurance and liability burdens could tilt away from companies that once employed human drivers and toward manufacturers of autonomous vehicles. The wildcard truly is how these vehicles will perform when introduced to our chaotically human roadways in large numbers.

Strong Incentives for Adoption

For many Americans, the idea of hopping in a vehicle with no human driver still seems farfetched. But for a wide variety of companies, the logic of pushing forward with this technology is irresistible.

The trucking industry will save enormously by taking human drivers out of the equation. Autonomous vehicles happen to brake much more efficiently than people — so much so that some observers believe driverless fleets will yield additional fuel economies of up to 10 percent, a significant savings in such a fuel-intensive sector.

The change could also solve another big problem for U.S. trucking — a shortage of up to 70,000 drivers. And while a brand-new autonomous truck would certainly be expensive, the self-driving technology company Otto is focused on retrofitting trucks to become autonomous at relatively modest costs on the order of $30,000.

Meanwhile, the prospect of running tandem lines of autonomous trucks on U.S. interstates (multiple tractor-trailers drafting off of each other in train-like fashion), greatly appeals to the industry due to the cost savings and efficiency gains that could occur. Tandem formations moving as a single unit would likely reduce the number of interactions — and therefore accidents or collisions — between trucks and non-professional drivers, thus slashing liability-related costs even further.

Preparing for the Shift

Given the momentum behind autonomous vehicles, it would be prudent for risk-managers, attorneys, C-suite executives and other professionals to start educating themselves on the various ways in which autonomous vehicles could affect what they do. The changes here could be profound.

Lawyers who run late-night TV ads to scare up plaintiffs for traffic accident cases could find themselves out of a job. Experts in accident reconstruction could become obsolete thanks to the robust, post-accident data that will be recoverable from driverless trucks and cars.

In all likelihood, our society will need a new army of tech-savvy experts who are ready to dive into complex litigation and other matters involving autonomous vehicles. They will need to fully understand the limits and capabilities of the systems in use and the laws and regulations in play. We might even see true technologists — trained engineers who know how to code — filling legal or risk-management roles.

Future cases could be extraordinarily complex: Two autonomous vehicles get into an accident in which a pedestrian is killed. Which vehicle was at fault? A hacker gains control of an autonomous vehicle and causes it to crash. Is the company liable or the manufacturer? A driverless shuttle drops off an elderly woman in a bad neighborhood and she is promptly mugged. Should the company have known better?

In some cases, the accidents may be the result of some defect in a vehicle’s control system. The absence of a driver (or drivers) takes issues such as perception-reaction time and driver fatigue out of the mix and those issues will be litigated with much less frequency. Most cases would be litigated over which system failed and who is responsible for the failure. Lawyers will need to be software engineers as much as lawyers.

Now is the time for companies to consider how the shift to autonomous vehicles could create opportunities and challenges for their businesses and affect their overall liability and risk-management profiles. Those who ignore this trend risk letting their better-prepared competitors take the wheel — if you’ll forgive the soon-to-be-outdated reference.

Peter M. Hart is senior counsel in the San Francisco office of LeClairRyan, and co-leader of its transportation industry team.
The opinions expressed are those of the author(s) and do not necessarily reflect the views of the firm, its clients, or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.