This morning, a PhD colleague asked us if I we were willing to participate in an experiment with robots that his friend was running. Another colleague and I readily agreed, so after lunch we headed to the Bristol Robotics Laboratory, a giant warehouse-shaped building which sits behind our own monstrous (by British standards) faculty building.
Before the experiment, we were asked multiple-choice questions about our perception of robots. The experiment itself was simple enough. The robot was supposed to give us instructions about how to build a toy cart but the only robot-like thing he did was point to the parts. I suspect the experimenter was controlling the robot as it was not a "smart" (autonomous) one. So, it was more an experiment on humans than on robots. I made what I was thought was a mistake in assembling the car but I realized quickly enough and corrected it without seeking the robot's or the experimenter's help. Maybe the long-dormant engineer in me woke up and put himself to the task of building the cart, robot or no robot. So, I suspect I was a bad experiment subject because I didn't really pay much attention to the robot's instructions once I got a fair idea of the assignment. It turns out the robot had been programmed to deliberately mislead me as the experimenter wanted to gauge my reaction to being misled. But, the plan bombed in my case because I thought I had made a mistake. In any case, the robot seemed dumb and annoying to me and that reflected in my answers in the post-experiment questionnaire.
Do I need to fear robots? I feel some tasks are beyond the capabilities of robots, although I must confess I know next to nothing on the subject. Sure, you can programme robots, teach them to mimic humans, teach them to "learn" from their mistakes like humans do, and of course, all these are tremendous achievements of science. But, can robots have consciousness? Can a robot be aware that she is a robot? Can they emote themselves, not just mimic human emotions? Do they have the capacity for imagination and abstract reasoning? Both are vital in research as well as artistic endeavours. I am sure there is a wide literature that answers (or goes some way towards answering) some of the questions I have raised. But, since this is a barely-read blog, and not an academic article, and as I don't really care that much about robots, I can't be bothered to seek this literature out.
Before the experiment, we were asked multiple-choice questions about our perception of robots. The experiment itself was simple enough. The robot was supposed to give us instructions about how to build a toy cart but the only robot-like thing he did was point to the parts. I suspect the experimenter was controlling the robot as it was not a "smart" (autonomous) one. So, it was more an experiment on humans than on robots. I made what I was thought was a mistake in assembling the car but I realized quickly enough and corrected it without seeking the robot's or the experimenter's help. Maybe the long-dormant engineer in me woke up and put himself to the task of building the cart, robot or no robot. So, I suspect I was a bad experiment subject because I didn't really pay much attention to the robot's instructions once I got a fair idea of the assignment. It turns out the robot had been programmed to deliberately mislead me as the experimenter wanted to gauge my reaction to being misled. But, the plan bombed in my case because I thought I had made a mistake. In any case, the robot seemed dumb and annoying to me and that reflected in my answers in the post-experiment questionnaire.
Do I need to fear robots? I feel some tasks are beyond the capabilities of robots, although I must confess I know next to nothing on the subject. Sure, you can programme robots, teach them to mimic humans, teach them to "learn" from their mistakes like humans do, and of course, all these are tremendous achievements of science. But, can robots have consciousness? Can a robot be aware that she is a robot? Can they emote themselves, not just mimic human emotions? Do they have the capacity for imagination and abstract reasoning? Both are vital in research as well as artistic endeavours. I am sure there is a wide literature that answers (or goes some way towards answering) some of the questions I have raised. But, since this is a barely-read blog, and not an academic article, and as I don't really care that much about robots, I can't be bothered to seek this literature out.
No comments:
Post a Comment