Bryant Walker Smith, Associate Professor at the University of South Carolina School of Law, visited the College of Law on October 28, 2019. In his presentation entitled “Automated Vehicles and Trustworthy Companies,” Professor Walker Smith spoke about the future of automated driving, how he believes companies will dominate this initiative, and the implications this will have on corporate trustworthiness. In discussing these topics, Walker-Smith sought to answer the question of who is driving driverless cars?
What is Automated Driving?
Walker Smith began by explaining the basics of automated driving and the development of automated vehicles. The law provides a generalized definition of driving, exemplified by the Maryland code which states that “driving means to drive, operate, move, or be in actual physical control of a vehicle.” MD Transp. Code §11-114, 141 (2016). This definition is fairly ambiguous, especially when considering the statute also states that “operate means to drive” but “operating is generally given a broader meaning than driving.” The ambiguities of the Maryland statute, and many others, can lead to problems when states inevitably have to adapt their traffic laws to account for automated driving and vehicles.
That being said, automated driving encompasses a diverse set of technologies that are ever changing, and this is a topic so many car companies and other innovators have been talking about and promising for years. There has been significant progress in this technological space, as more and more car manufactures are developing new models with increasingly automated features. However, these are not what most people would consider to be automated vehicles because they are not fully operable without human intervention.
The Society of Automobiles and Engineers defines six classifications of automated vehicles, and most new cars have features that would amount to Level 2 or Level 3 automation. At these levels, the human driver is primarily responsible for the operation of the vehicle but is aided by automated features such as adaptive cruise control and lane centering. Level 4 automation requires intervention of a human operator at some points, but is fully automated for the most part. Level 5 and Level 6 automation do not require a human operator.
Walker Smith spoke about two well-known examples of automated driving technologies are those being developed by Tesla and Uber. Tesla recently released their Smart Summon feature, which enables owners to summon their cars from another location in a parking lot and the car will navigate through the lot to the owner, all without a driver. Uber has also been developing their automated vehicle technology for years. They made the news when one of their vehicles struck and killed a pedestrian, even with a human operator also in the car. This crash was problematic because the human operator was distracted while in the vehicle and was completely reliant on the notion that the technology would not fail. While Uber’s attempt at developing automated driving technology was clearly unsuccessful, it raised important questions about how safety will be integrated into these technologies, even when a human operator is also involved.
The tragic accident involving Uber’s automated vehicle exemplifies that automated vehicles are still susceptible to error and can be equally as dangerous as human-driven cars. This accident brings to light that safety assurances need to be at the forefront of the further development of these technologies. However, the determination of whether a company has developed safe technology is not a one-dimensional inquiry. Walker Smith proposes that there are two main ways to test the safety of companies developing automated vehicle technologies: prospective safety and retrospective safety.
Prospective safety involves a general analysis of whether the public has reasonable confidence that the developer of the automated vehicle technology is trustworthy. Retrospective safety is more complicated and involves an analysis of whether the new automated vehicle technology is (1) at least as safe as a human driver, (2) at least as safe as any comparable automated vehicles, and (3) safer than the last automated vehicle that failed. While these two analyses are distinct, they are also intertwined as retrospective safety may naturally inform prospective safety.
Companies, Automated Vehicles, and the Law
Using the standards for analyzing safety as a framework, Walker Smith then transitioned to speak about how the reliability of automated vehicles will be determined by the reliability of the companies developing them, because the companies will play such an integral part in the development of these technologies. In order for these automated vehicle technologies to be successful, the corporate drivers must be trustworthy. A trustworthy corporate driver is one who will share its safety philosophy, make a promise of safety and reliability to the public, and keep that promise.
In practice, trustworthiness encompasses regulating companies and not the technology, while also requiring companies to vouch for their technologies and holding them accountable for any breaches in public trust. Sharing safety philosophies simply requires companies to outline their safety plan, why they believe that plan is reasonable, and why consumers should believe their statements. Their promises to the public would require that they only are marketing what is safe, are being candid about any limitations and failures of their technology, and rectifying any mistakes or failures related to safety or otherwise. Keeping these promises would be simple, requiring only that the companies manage public expectations, supervise their technology, and actively and publicly mitigate any harms their technology may cause.
This trustworthiness, in part, can be guided by both regulatory and tort law. Beginning with regulatory law, Walker Smith first discussed the Uniform Automated Operation of Vehicles Act, which serves as model legislation for the regulation of automated vehicles. This Act proposes that automated driving providers identify themselves to the United States government and represent that their vehicle are capable of complying with any and all relevant codes. Therefore, the company developing or providing automated vehicle technology can in essence be said to be the legal driver, that is until a human driver intentionally terminates the automation.
However, just like the technological developers need room to innovate in this industry, the same can be said for regulators given that this is a new area of technology with such a high risk for liability. To be effective, the regulation of this industry will need to focus on the processes and systems these companies are implementing to control their technologies. At the heart of this regulation is the desire to prevent breaches of public trust, which would include making hyperbolic claims, misrepresenting evidence, and failing to update products to address safety concerns.
The applicability of tort law to this industry also will help to address these regulatory goals. Breaches of public trust can also include exploiting litigation or forcing confidential settlements, which directly result from any tort liabilities the companies may have incurred. Imposing liability for automated vehicles will require these companies to integrate common carrier liability, strict products liability, and a quasi-fiduciary duty into their business models. Walker Smith proposes that this would impose four primary duties on providers of automated vehicle technology: the duty to refrain from puffing, the duty to update their technologies, the duty to promptly and fully compensate, and the duty to refrain from concealment. Imposing these duties would entail expansion of existing legal principles and doctrines to increase the burden imposed on companies developing automated vehicle technology.
Walker Smith concluded by again asking a question he posed several times throughout his presentation: who drives automated vehicles? The answer to this question is clearly companies, and the companies developing automated vehicles have significant power. However, this power is not without burden as the companies have a significant responsibility to protect and ensure the safety of the public as they develop these cutting-edge technologies.