Mrs. Johnson stares out the window. She keeps hoping her children will show up for an unexpected visit, although that rarely happens nowadays. Some days, Mrs. Johnson only sees Jenny from home care. And to Mrs. Johnson’s regret, Jenny never has time for a cup of coffee. Her children, all of whom try to balance a 40-hour work week and raising children, usually visit her twice a month on Sundays, if they’re not too busy that is. She cannot help but feel neglected, especially because she was always there for the kids when they were little. “Mrs. Johnson, would you like some coffee?” a voice speaks. Mrs. Johnson looks up – where did that voice come from? A small, human-looking robot approaches her. Mrs. Johnson closes her eyes, folds her hands and prays the creature will be gone by the time she opens her eyes again…
When I tell people I do research on healthcare robots, I often get negative reactions. Many people are appalled by the idea of their (grand)parents being taken care of by a robot: “What a ridiculous idea! Robots for physical labor are okay, but you really need a human for social tasks. There are plenty of unemployed people!!!” Why do robots instill such negative reactions? Are we afraid that they will take our place, as is often suggested in newspaper headlines (e.g., “Will Robots Take Your Job?” and “The cute robot that may eventually take your job”)? Or is there something else going on? In this blog, I explain why we are afraid and shed some light on how (un)realistic this fear of robots actually is.
One of the explanations of our fear of robots is the so-called uncanny valley (Mori, 1970). According to Japanese scientist Masahiro Mori, we get an uncanny or eerie feeling when robots look a bit like humans, but do not look and behave human-like enough to be assessed as human. Suppose that you think a person is standing a few meters away. After looking at the person for a while, his/her movements appear unnatural and start to creep you out. Only when you get closer you figure out that the person was in fact a robot. This is exactly what the uncanny valley entails; the robot looks human enough from a distance, but creates an uncanny feeling because of its unnatural behavior. To avoid these feelings, robots need to be designed such that their looks and behavior match: The robot is either approached as human, or clearly approached as a robot (and not somewhere in between).
Another explanation for our fear of robots may be that we are afraid that they will one day take over, or that they will, somehow, evolve beyond our control. This concept is often referred to as singularity. It assumes that, once we build a robot that is more intelligent than the human kind, this robot will start to develop even better robots and artificial intelligence, at which point us humans no longer have control over those robots (Vinge, 1993). In the end, this could lead to dystopian scenarios such as those seen in the Matrix – where the machines use ‘hibernating’ humans as energy source – or I, Robot – where robots start locking people up as the robots have become overzealous in their job of protecting humans. Some scientists suspect that singularity will take place somewhere between 5 and 100 years from now, although other scientists, such as psychologist Steve Pinker, do not believe in the concept of singularity at all: “There is not the slightest reason to believe in a coming singularity. The fact that you can visualize a future in your imagination is not evidence that it is likely or even possible.”
So how realistic is this fear of robots? Most robots currently being developed are only able to do a single thing really well. For instance, Roomba is really good at vacuuming, but is unable to bring you a cold beverage on a sunny day. Robotic seal Paro is really cuddly, and invitating for social interaction (either with Paro, or about Paro with other people around you), but it cannot do much else. For instance, Paro cannot decide for itself to go to the neighbor because it is unable to move independently. And the Japanese robot Ri-man is great at lifting patients, but it will not remind you of your upcoming appointments or that you have to take your medication. Many different kinds of robots already exist, but thus far none of them can do all of the tasks we would want, need, or even fear that robots could do. This is due to the enormous complexities that are involved in the integration of different robotic systems. Social robots in particular seem to be hard to create, as most of these robots currently can only do pre-programmed interactions or are secretly controlled by a human (a research technique often referred to as Wizard of Oz).
“Would you like some coffee, Mrs. Johnson?” Mrs. Johnson turns around and recognizes her home robot Alice. It has only been a few weeks since Mrs. Johnson was first introduced to Alice, but she already grew quite fond of her mechanical companion. “Ah yes Alice, coffee would be greatly appreciated!” Mrs. Johnson replies. “You know what Alice”, she adds, “Just sit down, I will make some myself!”
Mori, M. (1970). The uncanny valley. Energy, 7(4): 33-35.
Vinge, V. (1993). The coming technological singularity: How to survive in the post-human era. Vision-21: Interdisciplinary science and engineering in the era of cyberspace, 11-22. Retrieved on October 26th, 2015 from: http://www-rohan.sdsu.edu/faculty/vinge/misc/singularity.html
 This is an actual message found online in reaction to a newspaper article about healthcare robots that appeared in the Dutch newspaper Metro in 2012.
Marloes Spekman works as a PhD candidate within the SELEMCA project. The SELEMCA project focuses on the use of technology, such as robots and virtual agents, in the health care domain. Within the project, Marloes specifically focuses on how people’s prior emotions affect their perceptions of healthcare robots.