Check if public engagement itself may be harmful
Let us continue our discussion of potential problems with studying the driverless car dilemma (should a self-driving car swerve and kill one pedestrian to save five others?). One possibility is that talking about these morbid dilemmas might scare people away from adopting potentially safer autonomous cars.
I was once on an AI ethics panel organized by a major car manufacturer. One of my co-panelists was a famous philosopher, who was not a big fan of the ethical dilemmas of self-driving cars. To him, the question of whether a self-driving car should swerve and kill one pedestrian to save five others was a distraction, at best. He argued that going all the way back to Thomas Aquinas, philosophers knew that no matter what you do in such situations, you have a morally defensible position, and that is the end of the discussion.
In the drinks session after the panel, the philosopher came up to me, cocktail in hand, and said: “Do you realize that talking about this dilemma will scare people away from self-driving cars, and this will lead to more deaths? You will have blood on your hands!” After overcoming my initial indignation, I thought about what he said and it got me worried. What if he’s right? Could our very attempt to understand people’s fears be increasing those fears?
So, together with my collaborators Azim Shariff and Jean-François Bonnefon, we conducted a study to explore this. We recruited participants who never heard of the ethical dilemma of self-driving cars. We asked them whether they would be willing to adopt self-driving cars. But we showed half of them the dilemma first. If confronting the dilemma causes people to be less enthusiastic about robocars, then the philosopher would be right: exposure to the dilemma itself has a detrimental societal effect. To our relief, we could not detect such an effect. People can be trusted with thinking about the ethics of new technologies. But the episode is a cautionary tale, about the moral minefield that one traverses while trying to figure out machine ethics.
References
Jean-François Bonnefon and Azim Shariff and Iyad Rahwan. The Moral Psychology of AI and the Ethical Opt-Out Problem. in Ethics of Artificial Intelligence (ed. Matthew Liao, S.) (Oxford University Press, 2020).