By
Mike Nelson and Trevor Satnick
June 15, 2017, 1:03 PM EDT
Law360, New York (June 15, 2017, 1:03 PM EDT) —
Mike NelsonTrevor SatnickIt is becoming clear that the disruptive nature of the autonomous vehicle will extend far beyond the automotive industry. Conversations with clients from a variety of industries have provided unique insight into what a future might look like without humans behind the wheel.
Adopting a Challenge
Autonomous vehicles are expected to cut accident rates by nearly 90 percent, lower emissions, bring down the costs of auto insurance, and eliminate traffic jams. However, the technology raises two pertinent questions: (1) how will liability be apportioned in the event of an accident; and (2) how quickly will consumers adopt and drive automated vehicles daily?
While a drastic decrease in accident rates is expected, accidents will still occur. In fully autonomous vehicles (National Highway Traffic Safety Administration level 5), human beings, acting merely as passengers, will not have a hand in causing an accident. In these instances, will occupants still be required to obtain automobile insurance or will equipment manufacturers and automakers apportion liability among themselves? Muddying the waters even further are level 3 and level 4 autonomous vehicles, which require some level of human involvement. Insurance companies will have to establish risk factors when various levels of human participation work in concert with the machine.
Driving has been woven into the fabric of U.S. culture for more than a century. Relinquishing control and actively placing the lives of vehicle occupants in the hands of a machine will be a daunting task for many during the autonomous vehicle maturation process. While millennials are more accepting of the new technology than those of older generations, there is still trepidation. Coupling this innate fear of the unknown with highly publicized accidents involving level 2 autonomous vehicles may cause widespread adoption to take longer than safety data suggests it should. With that said, it is estimated that more than 600,000 preorders have been placed for Tesla’s Model 3 sedan. The company’s “Autopilot” function (level 2 autonomous driving suite) has been determined to be the car’s most desirable upgrade. While still early in development, it is clear that there is an appetite for autonomous technology in vehicles.
To Insure or Not to Insure?
A shift is already underway where insurance companies set their premiums based on how one drives rather than on age, gender and location of the driver. As artificial intelligence takes the wheel, the auto insurance market will shift away from consumers and will begin offering products that focus on liability for design defects, protection against software licensing disputes, and challenges to the robustness of the artificial intelligence algorithm.
A bill introduced in the U.K. (the Vehicle Technology and Aviation Bill) looks to maintain the status quo when it comes to auto insurance by requiring autonomous vehicle owners to purchase and insure their own vehicles, even when the machine is doing the driving. The U.S. will likely follow suit. While this solution suffices for the time being, it will make less sense as the machine becomes more involved and requires less and less human intervention.
A Fork in the Road
Autonomous driving presents certain ethical issues. For example, an autonomous vehicle may have to choose between protecting its occupants and protecting pedestrians crossing the street. How is the vehicle supposed to weigh the value of a human life? Will an algorithm be allowed to undertake a cold calculation and perform a cost benefit analysis based strictly on liability? What if cars are programmed to avoid contact with pedestrians at all costs? What incentive will pedestrians have to wait for the proper traffic signal when crossing a street knowing that a vehicle will always yield? Those teaching an autonomous vehicle to “think for itself” must understand and plan for these types of scenarios.
Information is Power
As original equipment manufacturers (OEMs) begin constructing autonomous vehicles for consumers, those OEMs will gain access to a massive data supply. This massive data supply will give OEMs a significant advantage when it comes to monetizing data. However, the OEMs will still have to abide by privacy policies or risk facing legal action. It is important to keep an eye on how these organizations structure their privacy policies in an attempt to capitalize on this new source of information.
A Fleet at One’s Disposal
Not only will vehicle operation change, but vehicle purchasing will change as well. In fact, there may be no need to purchase a vehicle at all. Riders will likely be able to hail an autonomous vehicle similar to how riders hail an Uber or Lyft ride today. Of course, this will result in another significant shift in the auto insurance industry as consumers will no longer need to personally insure their own vehicles.
Vehicle fleets should also revolutionize urban and suburban planning. Vehicles consistently on the move from one rider to the next should eliminate the need for parking spaces close to a destination. If and when the car is no longer in use or needs to be refueled (or recharged), it can return to a “home base.” Large parking centers can be placed in areas where real estate is not at a premium allowing for a more efficient and less costly means of transportation.
Car Theft: A Remote Issue
Autonomous vehicles will be connected to the internet in some capacity. This connection will create vulnerabilities that may be exploited by hackers looking to break into vehicle systems remotely. While computer hacking has been on the rise in recent years, the issue becomes even more acute when dealing with autonomous vehicles. The need to source a variety of components and software from different parties increases the potential for exploitation of vulnerabilities.
As vehicles become more complex, recalls have been issued more frequently. While security patches can be applied remotely, auto manufacturers and OEMs will have to play a “cat and mouse” game with bad actors, similar to how computer software designers must constantly release updates to keep up with newly discovered vulnerabilities today.
A Diverse, Competitive marketplace
Patent filings show an extraordinarily diverse marketplace with no one OEM, automaker or software supplier dominating the autonomous vehicle industry. The unsettled marketplace is creating tremendous opportunities for growth, leading to an explosion of startups that have entered the autonomous vehicle space in various capacities.
Right Idea, Wrong Method
While autonomous vehicles and artificial intelligence have been advancing at a rapid clip as of late, the technology could have been implemented much sooner had the powers that be settled on the correct method of building artificial intelligence. For decades, it was thought that the proper method of computer programming involved feeding the computer all possible scenarios and allowing the computer to calculate the best outcome based on the circumstance. This method worked well when a machine attempted to solve a problem with a limited set of known variables, like a chess match. A computer can be programmed to know each and every possible move because there are only a limited number of moves that can be made on a chess board.
This method won out as the de facto means of programming robots and set back the development of artificial intelligence for decades. In contrast, the machine learning method — allowing a robot to learn as it performs multiple trials in an attempt to achieve a goal — is far better at adapting in an unpredictable, real-world environment.
Looking Out for a Driving Companion
In pursuit of creating the mainstream autonomous vehicle, companies are partnering up in ways never seen before. Historically, companies headquartered in Detroit and Silicon Valley could not be more dissimilar. However, an autonomous vehicle requires highly advanced software to operate. Whether that software is designed around ride sharing or active vehicle management, automakers realize the importance of the software component and have sought out partnerships with technology companies working within the artificial intelligence and automation spaces.
Waymo (Google’s self-driving brand) has partnered with Fiat-Chrysler, GM has partnered with Lyft, and Volvo has partnered with Uber. It will be interesting to observe whether these partnerships will stand the test of time as the industry matures. Tesla originally partnered with Mobileye to develop its autonomous driving functionality, but the two companies parted ways. Tesla is currently developing its own solution in-house. Other automakers may follow suit once they gain the experience and assistance necessary to create their own proprietary systems.
Robots in Car Camouflage
A robot is defined as a “machine capable of carrying out a complex series of actions automatically.”[1] Many may not realize that fully autonomous vehicles will likely be the first robots that humans interact with on a daily basis. Hollywood would have us believe that a robot must look and act like a human, but the autonomous vehicle will actually become the first machine to fully perform a task without requiring any human input.
Legalizing Automation
Dramatic legislative changes are coming in the next few years to deal with automated and connected vehicles, artificial intelligence and robotics, and the use of an individual’s personal data. These laws will differ widely between countries. The U.S. has not issued any binding federal regulations on autonomous vehicles, leaving the individual states to determine how testing and implementation will be carried out. At least in the near-term, OEMs and auto manufacturers will need to contend with a patchwork of rules and regulations. Certain states, eyeing the economic boost associated with autonomous vehicle development, have been amenable to autonomous vehicle testing and integration, while other states are taking a “wait and see” approach. If the autonomous vehicle demonstrates its supposed safety and convenience benefits, it is likely that other states will follow suit and allow these new vehicles on the roadways.
Vehicle automation, artificial intelligence and robotics are still relatively nascent fields. Therefore, OEMs, tech providers and intermediaries within the ecosystem must stay on top of legal changes and factor the impacts into their technology builds, business models, customer documentation and contracts with counterparties. There is little regulatory guidance on apportionment of risk. In turn, contracts with counterparties should fairly allocate responsibility and risk with respect to (among other things) compliance with rules around personally identifiable information, artificial intelligence outputs and responsibility, intellectual property ownership, cyber threats and reporting, product liability, and software bugs.
More to Come
These takeaways only begin to scratch the surface when it comes to the kinds of changes automation will bring. As the technology continues to mature, its reach will spread to other industries that once appeared insulated. Like other technologies that started small and grew exponentially over time, automation and autonomous vehicles in particular appear primed to explode onto the scene.
Yet, it is more likely we will not even notice the shift occurring right under our noses. One day soon, while being chauffeured around in a vehicle without a steering wheel, we will look back and wonder how we ever survived without our first robots.
Michael R. Nelson is a partner and Trevor J. Satnick is a data privacy and security consultant at Eversheds Sutherland in New York.
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.
