United States: Liability Risks Of Automation And Connectivity In A Technologically Advanced World

1 Introduction

Advancements in technology and automation have led to an increasingly safer, convenient, and connected world. A few areas greatly impacted by these new technologies are autonomous vehicles, manufacturing, and the Internet of Things. While these advancements offer numerous benefits, they also can expose manufacturers, distributors, and insurers to new and potentially greater product liability risks. This article will discuss these advancements and the potential legal exposures they create.

2 Autonomous Motor Vehicles

The most prominent example of increased automation, and the one most likely to have the biggest effect on society, is the incorporation of increasing levels of automation in motor vehicles. Motor vehicles have become safer since their invention in the early 20th century. Seat belts, airbags, mirrors, indicator lights, anti-lock brakes, children's car seats, Bluetooth, power steering, and other features are taken for granted now, but were at one time pioneering safety features. However, in 2015, motor vehicle accidents were still responsible for the death of more than 35,000 people in the United States. See NHTSA, 2015 Motor Vehicle Crashes: Overview, Report No. DOT HS 812 318 (August 2016). The overwhelming majority of these accidents are caused by human error, including drunken driving, poor judgment, poor driving skills, poor reflexes, inattentiveness, poor vision, or criminal negligence. See Self-Driving Cars and Insurance, Insurance Information Institute, July 2016 (http://www.iii.org/issue-update/self-driving-cars-and-insurance). A staggering 94 per cent of automobile accidents are caused by human errors. See Critical Reasons for Crashes Investigated in the National Motor Vehicle Crash Causation Survey,National Highway Safety Traffic Administration (February 2015) (https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812115).

A study by the Insurance Institute for Highway Safety (IIHS) concluded that improvements in design and safety technology have led to a lower fatality rate in accidents involving late-model cars. See Self-Driving Cars and Insurance, Insurance Information Institute, July 2016 (http://www.iii.org/issue-update/self-driving-cars-and-insurance). The likelihood of a driver dying in a crash of a late-model vehicle decreased by more than a third over three years, and nine car models had zero fatalities per million registered vehicles. Among the factors to which the IIHS has attributed the lower death rate are the adoption of electronic stability control, which has reduced the risk of rollovers, and side airbags and structural changes that have improved occupant safety.

One way for motor vehicles to be even safer is through the increased use of autonomous motor vehicles. Autonomous driving will not only likely lead to significantly safer driving, but also provide greater mobility for people who are currently unable to drive, e.g., handicapped and elderly persons. Many people do not realise the extent to which automation has already improved driver safety. One example is electronic stability control systems, which help drivers maintain control while driving on slippery surfaces.

Motor Vehicle Manufacturer Innovations

Some car manufacturers have incorporated "driver assistance" packages in many of their models. These packages include parking systems with a rear-view camera, settings that allow for different driving modes based on weather and road conditions, side-assist blind-spot monitoring, dynamic variable ratio steering, adaptive cruise control, active lane departure warning systems, and top and corner camera view systems.

Forward sensor systems on several types of vehicles can detect if a forward collision is imminent and engage the automatic braking system at multiple levels without driver assistance to reduce the likelihood of a forward collision. If necessary, these systems can also activate protective measures, e.g., the front seat belts are pre-tensioned electronically, and the windows and sunroof close. Other systems allow the autonomous motor vehicles to analyse driving behaviour and warn drivers if they find any indication that drivers are starting to lose concentration. The systems use data from the radar and ultrasound sensors and a front camera, and guide cars by using gentle steering interventions that enable them to follow the line of vehicles ahead. These systems use roadway markings and other vehicles on the road to orient accordingly.

Preventive measures in certain systems protect against rear-end collisions. While the autonomous motor vehicles are in reverse, these systems assess the surrounding traffic and warn the driver of vehicles whose approach they deem critical. The systems have different levels of warning: visual; acoustic; and, finally, a short jolt of the brakes.

Several car manufacturers are incorporating these features in higher-end vehicles, and some are incorporating them in slightly less expensive models. Over time, the hope is that these features will be the new standard, i.e., they will be as common and required like seat belts. Forecasts predict that there will be 10 million automated vehicles on public roadways by 2020. See "10 Million Self-Driving Cars Will be on the Road by 2020," John Greenough, Business Insider (July 29, 2015) (http://www.businessinsider.com/report-10-million-self-driving-cars-will-be-on-the-road-by-2020-2015-5-6).

While automation can be used within a solitary vehicle to make driving safer, it can also be used among multiple vehicles to share information that can help mitigate the likelihood of an accident. The U.S. Department of Transportation has proposed a rule that would require motor vehicle manufacturers to include vehicle-to-vehicle communication technologies, thereby enabling several new crash-avoidance applications that could prevent "hundreds of thousands of crashes every year by helping vehicles 'talk' to each other". See U.S. DOT advances deployment of Connected Vehicle Technology to prevent hundreds of thousands of crashes (December 13, 2016) (https://www.nhtsa.gov/press-releases/us-dot-advances-deployment-connected-vehicle-technology-prevent-hundreds-thousands). The rule would require the devices to communicate using the same "language" through standardised messaging within the motor vehicle industry.

A relatively sizable obstacle to the increased use of autonomous vehicles is the United States' unique legal and legislative framework. The U.S. system comprises 50 states, with individual laws, and governed in certain circumstances by federal law. States' laws are often inconsistent with each other and/or contradictory. The National Highway Traffic Safety Administration (NHTSA) realised that it is problematic for international vehicle manufacturers to be governed by disparate regulations and published a set of guidelines intended to bring more uniformity to the manufacture of autonomous vehicles. California, Nevada, Michigan, North Dakota, Tennessee, Florida, and the District of Columbia were the only jurisdictions as of the beginning of 2016 that had legislation regulating autonomous vehicles on public roadways. Even those regulations differ from each other.

NHTSA – Federal Automated Vehicles Policy – September 2016

The NHTSA realises that this legal and regulatory framework is a potential obstacle on the path to safer driving. Accordingly, its Federal Automated Vehicles Policy: Accelerating the Next Revolution in Roadway Safety (September 2016) provides guidelines to consider for federal and state authorities, and motor vehicle manufacturers.

The four main parts of the new policy are:

• 15-Point Safety Assessment: The Vehicle Performance Guidance for Automated Vehicles for manufacturers, developers and other organisations includes a 15-Point Safety Assessment for the safe design, development, testing, and deployment of automated vehicles. The 15-Point Safety Assessment would cover: (1) data recording and sharing; (2) privacy; (3) system safety; (4) vehicle cyber-security; (5) human-machine interface; (6) crashworthiness; (7) consumer education and training; (8) registration and certification; (9) post-crash behaviour; (10) federal, state, and local laws; (11) ethical considerations; (12) operational design domain; (13) object and event detection and response; (14) fall-back (minimal risk condition); and (15) validation methods.
• Model State Policy: Delineates the federal and state roles for the regulation of highly automated vehicle technologies as part of an effort to build a consistent national framework of laws to govern self-driving vehicles.
• Current NHTSA Regulations/Options for Expediting Introduction: Outlines options for the further use of current federal authorities to expedite the safe introduction of highly automated vehicles into the marketplace.
• Modern Regulations/Identifying and Removing Obstacles: Discusses new tools and authorities the federal government may need as the technology evolves and is deployed more widely.

The policy also adopted the SAE International definitions for levels of autonomous driving. Those levels are as follows:

The goal of the policy is to set forth a proactive safety approach that provides life-saving technologies for motor vehicle operators while also allowing room for companies to innovate to develop new solutions. The U.S. Department of Transportation (DOT) believes that autonomous vehicles will provide "enormous potential benefits for safety, mobility, and sustainability".

3 Increased Automation in Other Industries

The effects of automation are not limited to motor vehicles. Automation is a factor in other emerging disrupting technologies such as the Internet of Things, drones, artificial intelligence, nanotechnologies, 3D printing, virtual reality, and blockchains. These technologies will transform life, business, and the economy. For example, while many politicians will use alternative facts to say that the United States has lost factory jobs because of trade, 88 per cent of factory jobs have been lost because of increased productivity via improvements in machinery and automation. See The Myth and the Reality of Manufacturing in America, Michael J. Hicks and Srikant Devaraj (June 2015) (http://conexus.cberdata.org/files/MfgReality.pdf). These job losses will not be confined to the past because the automation revolution is continuing. Recent studies from McKinsey and the economists Carl Benedikt Frey and Michael A. Osborne estimate that around 45 per cent of workers currently perform tasks that could be automated in the near future. See A Future That Works: Automation, Employment, and Productivity, McKinsey Global Institute (January 2017); and The Future of Employment: How Susceptible are Jobs to Computerisation?, Carl Benedikt Frey and Michael A. Osborne, September 17, 2013. The World Bank estimates that around 57 per cent of jobs could be automated within the next 20 years. In December 2016, the White House released a report wherein experts predicted that in the next 10 to 20 years, 47 per cent of jobs performed by humans in the United States could be replaced by advances in automation. See Artificial Intelligence, Automation, and the Economy, Executive Office of the President (December 2016) https://obamawhitehouse.archives.gov/sites/whitehouse.gov/files/documents/Artificial-Intelligence-Automation-Economy.PDF). These job losses will mostly affect jobs in the manufacturing, electronics, and pharmaceutical industries.

There are several reasons for the increase in automation, including increases in powerful and energy-efficient processors, open source software, and cheap sensors. These allow for "smart" autonomous products. See The Age of Autonomous Robots Is Upon Us, Fortune (March 29, 2016) (http://fortune.com/2016/03/29/autonomous-robots-startups/). Venture funding for robotics has grown to $922.7 million in 2015, up from $341.3 million in 2014. Projections indicate that the world will spend $135.4 billion on robotics and related services in 2019, up from $71 billion in 2015. See The Age of Autonomous Robots Is Upon Us, Fortune (March 29, 2016) (http://fortune.com/2016/03/29/autonomous-robots-startups/). As one might guess, this has led to an increase in desirability among graduates who might have otherwise pursued jobs in other industries. The effects of increased automation can be seen not only on factory floors and in manufacturing plants, but in small businesses, residences, and offices. Business owners, including retail giants like Amazon, use robots to monitor and stock shelves. In the near future, one can expect deliveries to be performed by drones and/or robots.

Homeowners can use a Roomba®, or similar products, to vacuum houses and robots to mow lawns. Consumer electronics, pool systems, plumbing systems, alarm systems, air conditioning and heating, security systems, sprinkler systems, washers, dryers, dishwashers, etc. can all be automated and/or operated remotely.

4 The Internet of Things

A major component of incorporating automation in products is the ability to connect with other technology, networks, and objects that are not part of the tangible product. New technology, and society's increasing willingness to embrace and rely on it, will have far-reaching consequences for manufacturers' product liability exposure, and accordingly, insurance companies that underwrite the policies that insure against those risks. Society is increasingly connected by software and hardware. The principal force behind that interconnectivity is the Internet of Things. The Internet of Things is the inter-networking of physical devices, vehicles, buildings, etc. embedded with electronics, software, sensors, and network connectivity that enable these objects to collect and exchange data. Several reports and experts estimate that the Internet of Things will consist of almost 50 billion objects by 2020.

The Internet of Things is relatively new, so laws, rules, regulations, and jurisprudence will necessarily evolve with the technology. However, that will not stop its continued growth. The growth of the Internet of Things will increase convenience and connectivity, but it will come with associated risks. Some of those risks will be relatively predictable, but some have assuredly not even been contemplated.

A 2014 Goldman Sachs report identified five key applications in which the use of the Internet of Things is vital: wearables (e.g., Fitbits); connected homes; connected cities; connected motor vehicles; and the industrial internet (including transportation, oil and gas and health care). See The Internet of Things: Making Sense of the next mega-trend, Goldman Sachs IoT Primer (September 3, 2014) (http://www.goldmansachs.com/our-thinking/outlook/internet-of-things/iot-report.pdf).

5 Product Liability Automation Risks

While there are safety benefits, and likely revenue benefits, for manufacturers who rely on automation, there will also be the potential for increased liability exposure. For example, motor vehicle manufacturers have long been subject to product liability lawsuits. However, the number and types of product liability lawsuits will likely increase because of autonomous vehicles. Autonomous vehicle manufacturers, and manufacturers in general, will likely face product liability claims based on various theories of liability, including, inter alia, strict liability, negligence, and breach of warranty. Manufacturers of autonomous vehicles will not be the only entities that will have to worry about an increased exposure to product liability lawsuits. Manufacturers of the various component parts, e.g. software and hardware manufacturers, will also be exposed to greater product liability risk.

Assessing liability with autonomous motor vehicles will be complicated by the possible combinations of driving modes, ranging from no autonomy to full autonomy. Potential factors include the comparative negligence of a human operator's actions, the functionality of software and sensors manufactured by potentially dozens of legally separate entities, the designs of the autonomous driving systems, and the training and/or warnings associated with operating an autonomous vehicle, among others.

These liability calculations and apportionments will involve determining which of the many component part manufacturers, if any, played a role in an incident. As software algorithms become more vital to the success and failure of autonomous vehicles, motor vehicle manufacturers will necessarily have to be more focused on the integration of software and hardware. In product liability practice, it is well-established that apportionment of liability can be apportioned up and/or down the supply chain to the cause of a particular failure. These risks are often addressed between component parts suppliers and manufacturers under the terms of supply agreements where a contractual duty to defend and indemnify against damages caused by a malfunctioning device is delineated.

In a technologically advanced setting, apportioning liability will not be as straightforward as it used to be. There will certainly be novel and challenging legal issues. For example, imagine a relatively simple traffic accident between two cars. In the pre-automation era, the liability would most likely be apportioned between the drivers of the two automobiles, and potentially one of the two vehicles if there was a manufacturing defect. On the contrary, imagine an Uber-owned Toyota being driven autonomously and the other car being an autonomously driven Audi. There could be several unrelated injured individuals in the shared Uber. One of the issues will be whether either the Uber or the Audi was taken over by a human driver at the time of the accident. Another issue is whether the hardware and/or software in the vehicles were defective. Did Uber own the vehicle or did it lease the car to a local business and/or individual operating the vehicle? Was either vehicle hacked, thereby potentially exposing to liability the entities responsible for each vehicle's cyber-security? Are autonomous vehicles allowed in that jurisdiction? Was the specific level of autonomous driving in use at the time of the accident allowed in that jurisdiction? Was the driver contractually obligated to have his/her hands on the steering wheel even though the car was in an autonomous mode? Did the drivers of the autonomous vehicles update the vehicles' software to ensure that they were being driven with the most recent software? Did the vehicle manufacturers adequately warn the drivers/owners of the vehicles that the software needed to be updated? Did either vehicle have equipment that was not manufactured by the original equipment manufacturer? If so, depending on the law of that jurisdiction, was that reasonably foreseeable? Liability could potentially be apportioned to various points along the supply chain – and this is a simple hypothetical two-car accident. Imagine the issues extrapolated to an interstate pile-up.

However, that same automation and interconnectivity could also provide a clearer picture to develop via the vehicles' internal software, so it should theoretically be easier to discover what contributed to the accident than in an accident wherein there could potentially only be testimony from the two drivers.

A few examples of other potential product failures connected with increased automation include the potential for software and/or hardware glitches leading to frozen pipes causing water damage; a sprinkler system malfunction causing water damage; a remote security system failing, leading to a home invasion and/or an abduction; a "smart" consumer electronics system causing a fire; and gas/fuel leaks leading to fires and/or explosions. While the consequences of those may be tragic, they would be confined to residences and those therein. There are potentially even greater consequences when those losses are extrapolated to factories, hospitals, medical facilities, offices, buildings, etc. Those losses could not only cause personal injury and property damage, but also business interruption losses. As machines grow more autonomous, potential failures become more complex and difficult to diagnose. Moreover, because automated processes likely run more efficiently and quickly than manual processes, they are usually more integrated into other operations around a facility and sometimes even beyond that facility. Any failure predictably would increase the cost of that failure and have exponential effects the longer it went undiagnosed.

The increased connectivity via the Internet of Things also poses interesting post-sale considerations and potential responsibilities on manufacturers. A manufacturer's duties to warn at the time of sale are well established. However, if a manufacturer discovers new risks after a sale, the legal framework for a manufacturer's responsibilities is not nearly as established. The Restatement (Third) of Torts, published in 1998, included a post-sale failure to warn duty. However, only some states have adopted that stricter standard. Due to the increased connectivity, manufacturers will likely be held to a higher standard regarding notice of certain failures and notifying consumers/users of those failures.

Once manufacturers become aware of potentially risky software programs and/or product defect issues, they will need to act quickly to provide upgrades and/or recall the defective products. Product recalls are relatively common for certain types of products and are usually handled by various regulatory agencies. Due to the increased connectivity, and access to information regarding products and consumers' information, manufacturers and retailers will have fewer defenses for not recalling almost all products.

A different type of failure arises in connection with cyber-security issues. The necessary reliance on the Internet of Things and increased automation allows for the potential for servers and/or products to be hacked. Users' personal data could be accessed and stolen. If these issues are prevalent with a certain product, manufacturers, including those of the component parts and/or security systems contained therein, would likely be subject to product liability lawsuits and/or class action litigation. The investigations attendant to litigation will also require the use of novel and educated experts in those fields. Since this will be a new area of litigation, identifying the relevant experts will be a significant undertaking.

There are potentially bigger issues when outside parties hack devices connected to the Internet of Things for nefarious, political, and/or criminal motives. The U.S. Federal Trade Commission (FTC) issued a report in January 2015 that highlights these issues. See Internet of Things – Privacy & Security in a Connected World, FTC Staff Report (January 2015) (https://www.ftc.gov/system/files/documents/reports/federal-trade-commission-staff-report-november-2013-workshop-entitled-internet-things-privacy/150127iotrpt.pdf). The FTC report noted security concerns for consumers using devices connected to the Internet of Things, such as enabling unauthorised access and misuse of personal identification, facilitating attacks on other systems and creating safety risks. The report noted that while these risks exist with traditional computers and computer networks, they are more prevalent due the increased connectivity associated with the Internet of Things. Moreover, computers and computer networks have decades of experience in cyber-security. Many companies that have never before considered cyber-security in relation to their products will not have decades of that institutionalised cyber-security knowledge on which to rely.

6 Effect on Insurers

Almost all manufacturers have product liability insurance. Accordingly, insurers will have to consider all of the potential benefits and risks of automation when drafting policies and/or agreements with potential insureds. Insurers will also need to hire claims representatives familiar with the technology incorporated in the respective products, leading to a shift in the types of people employed by insurance companies and/or the qualifications necessary to work at insurance companies.

The framework for insurance will also shift, especially with the increased use of autonomous motor vehicles. The paradigm will likely shift from a user error focused evaluation to a product liability focus. As products become increasingly automated, the burden might be on the manufacturer to prove it was not responsible for an incident. This will be a marked shift from the old paradigm in motor vehicle accident evaluations. There is a small possibility, if the integration of autonomous motor vehicles is not seamless, that the liability issues could threaten the financial viability of motor vehicle manufacturers. However, it is likely that such a scenario would be prevented via regulation and/or legislative action.

Another consideration is that as consumers have more options for transportation, especially in public transportation, autonomous motor vehicles, and/or shared rides, car-ownership might decrease significantly. This could impact the types of insurance available and insurers' financials.

7 Conclusion

While advancements in automation may provide new risks and product liability considerations, they will also lead to dramatic increases in safety. It is possible that federal and state legislatures will enact legislation protecting manufacturers from the attendant risks and legal exposure. However, at least in the interim, those advances in safety will not be sacrificed because they may lead to more product liability exposure, so manufacturers, retailers, and insurers should be prepared for the new automation and product liability landscape and be willing to evolve as the technology around us evolves.

Originally published by International Comparative Legal Guide: Product Liability 2017, May 21, 2017.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.