Introduction

Self-driving cars may have seemed like the future decades ago, but advancements in technology have enabled testing of fully autonomous vehicles. These new vehicles bring many promises of benefits to society. However, designers must examine unintended ethical consequences to these benefits, along with ethical considerations to their implementation. While fully autonomous vehicles are not yet here, there is a great need to plan for the new ethical dilemmas they bring.

Kohl et al. state that current surveys show 56% of people have positive opinions of self-driving cars, while 13.8% have negative concerns, and 29.4% have a neutral opinion, noting a general acceptance of self-driving cars in the public (2017). Their own survey of Twitter data, however, showed 76.4% neutral tweets, with more discussions about the risks than the advantages (Kohl et al., 2017). The public has good reason to question this new technology, as there are still complex ethical questions to answer.

The Society of Automotive Engineers (SAE) lists five different levels of autonomous driving, with Level 1 implementing driver assistance, Level 2 using partial automation or multiple combined Level 1 systems, Level 3 with conditional automation, Level 4 having high automation, and Level 5 implementing complete automation (Groshen et al., 2018; Faggella, 2020). Most car company executives refer to vehicles above Level 3 as self-driving (Faggella, 2020). Current levels of self-driving vehicles are at Level 2 automation (Kohl et al., 2017). Even at this level, there are ethical complexities.

Autonomous Vehicle Ethical Pros and Cons

There are many ethical reasons to implement autonomous vehicles. The most important reason for implementation is the high levels of safety they would bring to roadways. Researchers found that at least 90% of traffic accidents are from human error, which self-driving cars could eliminate (Kohl et al., 2017; Fleetwood, 2017). Interconnected autonomous cars could share important information between them, allowing a more efficient distribution with higher number of vehicles moving faster and safer than traditional cars (Coco-Vila, 2017). Fleetwood believes that, with autonomous vehicles, “passengers have access to a level of safety and convenience that is unparalleled in other forms of transportation” (2017). Groshen et al. state that truckers could be safer and more comfortable by automation technologies like Advanced Driver-Assistance Systems, which are equivalent to Level 1 or 2 automation (2018). The act of “platooning,” trucks moving in a convoy controlled by a lead truck, also helps with safety and comfort of truck drivers (Groshen et al., 2018).

These safety features of self-driving cars, however, come with the loss of personal control of mobility and privacy. Many recent surveys have shown that while there is a general acceptance of self-driving cars, people may start to dislike the loss of control similar to when cruise control became available (Kohl et al., 2017). Boeglin believes that some will even find it dehumanizing (2015). Boeglin also notes that some vehicles may remove a driver’s autonomy of choosing their own route (2015). Enhanced safety may also require sharing details about the car with many different entities (Borenstein et al., 2017; Boeglin, 2015). California currently requires that autonomous vehicles need to store records of the 30 seconds leading up to a crash (Boeglin, 2015).

There are environmental positives from self-driving vehicles, and helping the environment can be an ethical priority. Currently, in the United States, the trucking industry provides 28% of supply chain transportation, but is responsible for 71% of the food supply chain’s emissions (Heard et al., 2018). Using automation technologies like platooning can help curb these emissions from reduced aerodynamic drag (Heard et al., 2018). Other autonomous vehicle features like optimal driving cycles, optimal routing and dynamic eco-routing, less idling, reducing cold starts, trip smoothing and harmonization, and lighter vehicles can have significant environmental emission reduction, especially for refrigeration trucks (Heard et al., 2018).

Groshen et al. state that automation will have economic advantages, with new autonomous vehicle related jobs, many with better salaries than those they would displace (2018). They also see other improvements to the economy, stating that consumers will buy more non-transportation goods and services as they spend less on transportation costs and have more free time during travel (Groshen et al., 2018). Heard et al. believe that there may be increased profits in the food distribution industry from cost savings of automation (2018).

Many workers, however, will see a loss of jobs from the automation. Heard et al. call the loss of jobs in the trucking industry a “substantially negative economic effect of autonomous vehicle technology” (2018). They state that in 2016, the heavy and tractor-trailer truck driving industry employed 1.7 million Americans, with autonomous vehicles potentially displacing these workers (Heard et al., 2018). A side effect of this is that rest stops, lodging, and restaurants along the highway system may also suffer, as they depend on truck drivers as customers (Heard et al., 2018). Researchers also see a reduction of many other jobs, including “bus drivers, taxi, and other personal transport employees” and “automobile insurance adjusters, auto repair mechanics, (or) police patrol officers” (Groshen et al., 2018). The public will need to consider how to handle the ethical dilemma of these job losses, with many of these workers needing education and training for transitioning to new autonomous vehicle jobs or other industries.

An ethically moral obligation for implementing autonomous vehicles is the mobility of the disabled. Bradshaw-Martin & Easton state that many disabled feel “inequalities in access to independent mobility” for their inability to drive current cars, and that a major ethical problem exists in excluding the disabled from having this mobility (2014). They argue that cars should have design requirements so that everyone can use them, including those with even ailments like blindness or Alzheimer’s disease, and any exclusion arbitrarily and unnecessarily creates an injustice (Bradshaw-Martin & Easton, 2014).

An ingenious design in automation can help the disabled gain this mobility back. A brilliant example of this is Ralph Teetor, a blind inventor, who, at a young age, learned to drive a car with the guidance of an overhead line (Bradshaw-Martin & Easton, 2014). Ralph Teetor went on to become the creator of cruise control, one of the first steps in vehicle automation (Bradshaw-Martin & Easton, 2014). Currently, many manufacturers do not include the disabled in car design decisions because they see them as a “costly afterthought” and do not yet perceive them as a viable market (Bradshaw-Martin & Easton, 2014). Steven Mahan, head of the Santa Clara Valley Blind Center, also believes that “society needs to accept the automation before accepting disabled drivers” (Bradshaw-Martin & Easton, 2014).

There should be ethical considerations for teaching and implementing features for aging drivers, too. Yang & Coughlin state that older drivers are less likely to learn these new autonomous vehicle systems as fast as a younger digital generation does (2014). However, older drivers are not reluctant to use these new systems once shown the benefits, though they do take longer to learn them (Yang & Coughlin, 2014). Other considerations for teaching the new technology to aging drivers include systematic differences between drivers as a function of age, generational effects like progressive motorization and gender differences, the changing lifestyles of the elderly, and age-related afflictions of chronic and acute diseases (Yang & Coughlin, 2014). Designers could implement self-driving cars with technologies to assist the elderly with features like advanced navigation systems for aging drivers or in-vehicle health monitoring systems (Yang & Coughlin, 2014).

There are legal and liability concerns with developing autonomous vehicles as well. The technology must be “accepted by the public as unproblematic,” which requires a good liability legal framework (Bradshaw-Martin & Easton, 2014). Bradshaw-Martin & Easton argue that manufacturers could have integrated current cruise control systems, which include anti-lock brakes, stability control systems, and navigation aids, into autonomous vehicles as early as the late 1990s, but held off development because of legal and liability concerns (2014). The Vienna Convention, with over 70 ratifications, and the Geneva Convention, which the United States and United Kingdom follow, both state that a driver must be able to control the vehicle, excluding partially sighted or other types of disabled people from being able to use a fully automated car (Bradshaw-Martin & Easton, 2014). Bradshaw-Martin & Easton believe that there are ethical reasons to update these treaties (2014).

There are questions regarding who is liable for vehicles in accident situations. In examining if a self-driving vehicle is responsible for its accidents, Gless et al. state that autonomous vehicles do not meet the requirements of personhood (2016). United States law places responsibility on the “managerial agent with supervisory responsibility of the subject,” and German law is likely unable to blame the vehicle, as it cannot recognize and evaluate its past actions with a moral reference system (Gless et al., 2016). Current German and United States liability laws do have previous cases for guidance on programming a machine for harm or simple negligence in its design, but note that intelligent agents like an autonomous vehicle push these laws to their limits, and questions remain on when a human must take over a self-driving vehicle (Gless et al., 2016).

The Trolley Problem and Accident Algorithms

Many ethical, legal, and liability concerns with autonomous vehicles stem from how accident algorithms solve the trolley problem. Philosopher Philippa Foot, who originally conceived of the trolley problem, describes it as a situation where five men are working on a train track and one man is working on another section of track, with a runaway tram heading toward the five men (1967). The train operator or another observer must make a choice to either throw a switch to move the tram to the other track, killing only one person, or take no action and allow the train to kill the five men (Foot, 1967). Similar situations can arise with various ethical consequences, with Foot providing examples of a judge framing an innocent man when the guilty party is unknown, preventing rioters from violently attacking a community if they do not see justice for a crime, or a pilot making the choice to crash land a failing aircraft to a less populated area (1967). With self-driving cars, there are many potential trolley problems, and accident algorithms will need some type of ethical examination for determining who might live or die.

Coca-Vila has identified five different examples of potential trolley problems in autonomous vehicles (2017). The dilemmas include the choice of death of a pedestrian or the driver from brake failure, between hitting jaywalkers or an elderly bicyclist, the driver’s death or two children playing in an emergency lane, death of a pedestrian or motorcyclist following too close to the vehicle, or between two motorcyclists, one with a helmet and one without (Coco-Vila, 2017). Each situation provides unique questions legally and ethically. For example, algorithms need to consider if a helmetless motorcyclist will die in a collision and instead should hit the motorcyclist wearing a helmet, or if it is morally justifiable to consider age in a potential accident against the bicyclist, or if any level of culpability matters for the jaywalkers (Coco-Vila, 2017).

Coco-Vila examines a few different reasons an algorithm might choose someone to collide with in his accident situations. He describes different considerations based on ethical egoism with a selfish self-driving car, a utilitarian self-driving car acting on behalf of the best public interest, or algorithms based on the “criminal theory of justification” (2017). Overall, Coco-Vila believes that ethical egoism is unacceptable because the legal system and ethical good cannot use that as a model to protect the non-driver’s rights, and utilitarianism does not work because of the rights and duties of each individual or “free agent” (2017). Instead, he argues that the criminal theory of justification offers “the most complete, exhaustive, and well-founded battery of arguments for solving conflicts,” along with legal justifications for self-defense (Coco-Vila, 2017).

While Coco-Vila’s legal analysis of each situation can be specific to European laws, there are many relationships and implications regarding laws in the United States. For example, Coco-Vila points out that in his solution for the two motorcyclists, hitting the helmet-wearing motorcyclist is the best option, as not wearing a helmet does not justify death (2017). The motorcyclist with the helmet on could pursue civil action against the one without a helmet as that was the cause of the vehicle choosing him in the accident (Coco-Vila, 2017). Other unintended ethical consequences from accident algorithms avoiding a helmetless motorcyclist in these scenarios may encourage riders to stop wearing helmets at all, violating many state traffic laws (Trappl, 2016). These types of choices have even further implications. A vehicle may assume that a smaller car as a target in an imminent crash situation increases its driver’s survival odds, potentially affecting small car safety and sales (Trappl, 2016).

Some ethicists disagree with the analogy between autonomous vehicle crash algorithms and the trolley problem. While Nyholm & Smids agree that “automated vehicles need to be programmed for how to respond to situations where a collision is unavoidable,” they believe there are three ways that accident algorithms differ from the trolley problem (2016). First, the algorithm decisions are predesigned before the accident is taking place with more than one person making the decision instead of in real time (Nyholm & Smids, 2016). Second, Nyholm & Smids argue that there are different moral and legal considerations to actual autonomous vehicles outside of the hypothetical trolley problem, explaining as an example that the law does not actually permit people to switch a tram track for various reasons (2016). Third, they state that there are risks and uncertainties in the real world that accident algorithms are unable to predict, like a sudden head on collision, or knowing that hitting a pedestrian will kill the individual and not simply injure them (Nyholm & Smids, 2016).

Holstein & Dodig-Crnkovic also disagree about the trolley problem analogy, stating that using the analogy creates a “belief in perfect predictability of complex systems involving vehicles and humans,” and expectations that cars can and should make a choice between different people in an accident (2018). They point out that differentiating individuals based on an individual’s age, profession, gender, or social rank is unethical according to the German ethics commission for autonomous driving (Holstein & Dodig-Crnkovic, 2018). Using the trolley problem analogy also has an assumption that there are specific outcomes in the event of a decision, which is unlikely simply due to the quality of sensors or detection equipment, and vehicles might not distinguish between groups of people based on number but instead by space occupied (Holstein & Dodig-Crnkovic, 2018). Holstein & Dodig-Crnkovic also note that humans learn from their mistakes, while software takes time to update (2018).

There are problems simply automating any trolley problem solution with code, too. Millar states that the automation of ethical decision-making may take the car’s driver out of any decision process, removing moral autonomy and causing a morally problematic paternalism, which goes against established ethical norms (2015). Millar gives an example of a woman driver potentially choosing to sacrifice herself to save a child in a trolley problem situation, while an autonomous vehicle may take away that end-of-life decision from her (2015).

Frameworks for Ethical Testing

While validating that an autonomous vehicle performs ethically is a complex topic, researchers have proposed new and existing frameworks as guidance. Fleetwood believes that the Code of Ethics for Public Health, a widely cited framework for public health ethics, can provide a starting point for health advocates to advocate for the rights of individuals while also protecting public health with public input (2017). Public health officials can also ensure that communities can make informed decisions on if and how autonomous vehicles work on their roads, making sure manufacturers obtain community consent on their operation (Fleetwood, 2017).

Millar has a slightly different opinion, believing that having only ethicists determine accident algorithm designs sets the bar too high (2016). He believes that enough ethical decisions will arise that there will be common designs, allowing for standardized ethics evaluation tools and frameworks (Millar, 2016). Millar proposes five rules to create these types of frameworks (2016). These five guidelines are that ethics evaluation tools for robotics are created by users and designers together, that any ethical framework is user-centered, it will acknowledge user-robot relations psychology, it should satisfy the principles of human-robotics interaction (HRI) Code of Ethics, and will distinguish between acceptable and unacceptable design features (Millar, 2016). Millar conveniently provides his own detailed framework that satisfies these five rules (2016).

Conclusion

It is clear that there are many ethical pros and cons with society implementing self-driving cars. With increased safety and mobility comes the loss of control, privacy, and potentially personal autonomy. The technology may enable those without previous mobility to have free movement. New engineering jobs requiring new skills and training, along with potential environmental benefits, may see the loss of traditional jobs and industries. To help determine the best ethical outcomes, there needs to be properly vetted guidelines and frameworks for the new future of our roadways.

References

Boeglin, J. (2015). The costs of self-driving cars: Reconciling freedom and privacy with tort liability in autonomous vehicle regulation. Yale Journal of Law and Technology, 17(1).

Borenstein, J., Herkert, J. R., & Miller, K. W. (2017). Self-driving cars and engineering ethics: The need for a system level analysis. Science and Engineering Ethics. https://doi.org/10.1007/s11948-017-0006-0

Bradshaw-Martin, H., & Easton, C. (2014). Autonomous or ‘driverless’ cars and disability: A legal and ethical analysis. Web Journal of Current Legal Issues, 20(3).

Coca-Vila, I. (2017). Self-driving cars in dilemmatic situations: An approach based on the theory of justification in criminal law. Criminal Law and Philosophy, 12(1), 59-82. https://doi.org/10.1007/s11572-017-9411-3

Faggella, D. (2020). The self-driving car timeline: Predictions from the top 11 global automakers. Emerj. https://emerj.com/ai-adoption-timelines/self-driving-car-timeline-themselves-top-11-automakers/

Fleetwood, J. (2017). Public health, ethics, and autonomous vehicles. American Journal of Public Health, 107(4), 532-537. https://doi.org/10.2105/ajph.2016.303628

Foot, P. (1967). The problem of abortion and the doctrine of double effect. Oxford Review, 5, 5-15.

Gless, S., Silverman, E., & Weigend, T. (2016). If robots cause harm, who is to blame? Self-driving cars and criminal liability. New Criminal Law Review, 19(3), 412-436. https://doi.org/10.1525/nclr.2016.19.3.412

Groshen, E. L., Helper, S., MacDuffie, J. P., & Carson, C. (2018). Preparing U.S. workers and employers for an autonomous vehicle future. https://doi.org/10.17848/tr19-036

Heard, B. R., Taiebat, M., Xu, M., & Miller, S. A. (2018). Sustainability implications of connected and autonomous vehicles for the food supply chain. Resources, Conservation and Recycling, 128, 22-24. https://doi.org/10.1016/j.resconrec.2017.09.021

Holstein, T., & Dodig-Crnkovic, G. (2018). Avoiding the intrinsic unfairness of the trolley problem. Proceedings of the International Workshop on Software Fairness - FairWare ‘18. https://doi.org/10.1145/3194770.3194772

Kohl, C., Mostafa, D., Böhm, M., & Krcmar, H. (2017). Disruption of individual mobility ahead? A longitudinal study of risk and benefit perceptions of self-driving cars on Twitter.

Millar, J. (2016). An ethics evaluation tool for automating ethical decision-making in robots and self-driving cars. Applied Artificial Intelligence, 30(8), 787-809. https://doi.org/10.1080/08839514.2016.1229919

Nyholm, S., & Smids, J. (2016). The ethics of accident-algorithms for self-driving cars: An applied trolley problem? Ethical Theory and Moral Practice, 19(5), 1275-1289. https://doi.org/10.1007/s10677-016-9745-2

Trappl, R. (2016). Ethical systems for self-driving cars: An introduction. Applied Artificial Intelligence, 30(8), 745-747. https://doi.org/10.1080/08839514.2016.1229737

Yang, J., & Coughlin, J. F. (2014). In-vehicle technology for self-driving cars: Advantages and challenges for aging drivers. International Journal of Automotive Technology, 15(2), 333-340. https://doi.org/10.1007/s12239-014-0034-6