What would you do if you saw a self-driving car hit a person?
In Robot Ethics, Mark Coeckelbergh, professor of Philosophy of Media and Technology at the University of Vienna, poses a trolley problem for 2022: The car should continue its course and kill five pedestrians, or deviate from its course and kill one?
In the chapter presented here, Coeckelbergh examines how humans have conceptualized robots as a larger framework and looks at how self-driving cars will handle lethal traffic situations, and whether this is a worthwhile question.
In the 2004 American science fiction film I, Robot, humanoid robots serve humanity. But all is not well. After an accident, a robot rescues a man from a sinking car, but a twelve-year-old girl is not saved. The robot calculated that the man had a better chance of survival; humans may have made another choice. Later in the film, the robots try to take over the power of the humans. The robots are controlled by an artificial intelligence (AI), VIKI, who decided that limiting human behavior and killing some humans will ensure humanity’s survival. The film illustrates the fear of humanoid robots and AI taking over the world. It also points to hypothetical ethical dilemmas if robots and AI reach general intelligence. But is this what robot ethics is and should be about?
Are robots coming or are they already here?
Usually, when people think of robots, the first image that comes to mind is a highly intelligent, human-like robot. This image is often derived from science fiction, where we find robots that look and behave more or less like humans. Many narratives warn about robots. who take over; the fear is that they will no longer be our servants but that they will make us their slaves. The term “robot” itself means “forced labor” in Czech and appears in Karel Čapek’s play RUR, performed in Prague in 1921, just over a hundred years ago. The play is part of a long history of stories about rebellious human-like machines, from Mary Shelley’s Frankenstein to films like 2001: A Space Odyssey, Terminator, Blade Runner and I, Robot. In the public imagination, robots are often the object of both fear and fascination. We fear that they will take over, but at the same time it is exciting to think about creating an artificial being that is like us. Part of our romantic heritage, robots are projections of our dreams and nightmares about creating an artificial other.
First of all, these robots are mostly scary; they are monsters and strangers. But at the beginning of the 21st century, a different image of robots is emerging in the West: the robot as a companion, friend and perhaps even partner. The idea now is that robots don’t have to be confined to industrial factories or remote planets in space. In the contemporary imagination, they are freed from their dirty slave labor and enter the home as pleasant, helpful, and sometimes sexy social partners you can talk to. In some movies, they still rebel, think Ex Machina, for example, but they generally become what robot designers call “social robots.” They are designed for “natural” human-robot interaction, meaning interaction the way we’re used to interacting with other humans or pets. They are designed not to be scary or monstrous, but to be cute, helpful, entertaining, fun and seductive.
This brings us to real life. The robots are not coming; they are already here But they are not much like the robots we know in science fiction. They’re not like Frankenstein’s monster or the Terminator. They are industrial robots and sometimes “social robots”. The latter are not as intelligent as humans or their sci-fi relatives, however, and often lack human form. Even the sex robots aren’t as intelligent or conversational as the robot depicted in Ex Machina. Despite recent advances in AI, most robots are not human in any sense. That said, robots are here and they’re here to stay. They are smarter and more capable of operating autonomously than before. And there are more real-world applications. Robots are not only used in industry, but also in healthcare, transportation and home assistance.
This often makes life easier for humans. But there are also problems. Some robots can be downright dangerous, not because they’re trying to kill you or seduce you (although “killer drones” and sex robots are also on the menu for robot ethics), but usually for more mundane reasons like now because they can take your job. , can trick you into thinking it’s a person, and can cause accidents when you use it as a taxi. These fears are not science fiction; they refer to the near future. More generally, from the impact of nuclear, digital and other technologies on our lives and the planet, there is a growing awareness and recognition that technologies are making fundamental changes to our lives, societies and environment and therefore we had better think more and more critically, about its use and development. There is a sense of urgency: let’s better understand and evaluate technologies now, before it’s too late, that is, before they have impacts that no one wants. This argument can also be made for the development and use of robotics: we consider the ethical issues raised by robots and their use at the development stage rather than after the fact.
Autonomous cars, moral agency and responsibility
Imagine a self-driving car driving at high speed down a narrow lane. Children play in the street. The car has two options: either it avoids the children and crashes into a wall, probably killing the only human passenger, or it continues on its way and brakes, but probably too late to save the children’s lives. What should the car do? What will the cars do? How should the car be programmed?
This thought experiment is an example of the so-called trolley dilemma. A track cart is about to drive more than five people tied to a track. You are standing on the side of the track and can pull a lever that redirects the cart to another track, where a person is tied. Pull the lever? If you do nothing, they will kill five people. If you pull the lever, it will kill a person. This type of dilemma is often used to get people thinking about what they perceive to be the moral dilemmas posed by self-driving cars. The idea is that this data could help the machines decide.
For example, the online platform Moral Machine has collected millions of decisions from users around the world about their moral preferences in cases where a driver has to choose “the lesser of two evils.” People were asked whether a self-driving car should prioritize humans over pets, passengers over pedestrians, women over men, and so on. Interestingly, there are cross-cultural differences in the choices made. Some cultures, such as Japan and China, for example, were less likely to forgive the young than the old, while other cultures such as the United Kingdom and the United States were more likely to forgive the young. So this experiment not only offers a way to approach machine ethics, but also raises the more general question of how to account for cultural differences in robotics and automation.
Fig. 3 (Image: MIT Press/Mark Coeckelbergh)
Figure 3 shows an example of a cart dilemma situation: should the car continue on its course and kill five pedestrians, or deviate from its course and kill one? Applying the trolley dilemma to the case of self-driving cars may not be the best way to think. on the ethics of self-driving cars; fortunately, we rarely encounter such situations in transit, or the challenges may be more complicated and not involve binary choices, and this problem definition reflects a specific normative approach to ethics (consequentialism and, in particular, utilitarianism). There is discussion in the literature about the extent to which trolley dilemmas represent real ethical challenges. However, the cart dilemmas are often used to illustrate the idea that as robots become more autonomous, we have to think about whether or not to give them some kind of morality (if that can avoid at all) and if so, what morality. In addition, autonomous robots raise questions about moral responsibility. Think again about the self-driving car.
In March 2018, a self-driving Uber car killed a pedestrian in Tempe, Arizona. There was an operator in the car, but at the time of the accident the car was in autonomous mode. The pedestrian was walking outside the crosswalk. The Volvo SUV did not slow down as it approached the woman. This is not the only fatal accident reported. In 2016, for example, a Tesla Model S car on Autopilot failed to detect a large truck and trailer crossing the freeway and hit the trailer, killing the Tesla driver. For many observers, these accidents show not only the limitations of current technological development (currently cars do not seem ready to participate in traffic) and the need for regulation; they pose challenges regarding the attribution of responsibilities. Consider the Uber case. Who is responsible for the accident? The car cannot be held responsible. But all the human parties involved can potentially be held responsible: the Uber company, which wears a bandwagon hat, is not yet ready for the road; car manufacturer Volvo, which failed to develop a safe car; the operator of the car who did not react in time to stop the vehicle; the pedestrian who was not walking in the crosswalk; and the regulators (eg, the state of Arizona) that allowed this car to be tested on the road. How should we attribute and distribute responsibility given that the car was driving autonomously and there were so many parties involved? How we should attribute liability in all kinds of autonomous robot cases, and how we should approach this issue as a profession (eg engineers),…