Robo-Ethics

I went to a thing called Cafe Scientifique last night.  I happened upon one a month or so ago when it overlapped with the Skeptics In The Pub event that I was attending (only the second of those that I'd been to, as well!).  It struck me as rather a nifty idea - a person of a scientific persuasion shows up and talks about their specialty.  "Talks" being the active verb - it's something of a conversation with the group, not so much a lecture.

This week was a rather lively discussion on the ethics of Robotics, led by AJung Moon from UBC.  The conversation wandered a fair bit, and I'm afraid she didn't get a chance to talk about everything she'd planned to, but it was still fascinating and I'm glad she came out to talk with all of us.

There was a lot of discussion on defining what a Robot actually is.  Which I suppose shouldn't have been too surprising - a scientifically minded audience is going to want you to be precise with your definitions.  Perhaps a couple of people got overly into that sub-topic.

AJung did provide a definition - an entity that is capable of interacting with the world, through sensors and some sort of mechanism (IE: a camera and robot arm), and has some level of autonomous decision making capabilities.  It's a broad definition, but the topic of ethics should cover a wide range of situations (if not all!), so it seemed suitable to me.

The very breadth of the topic is a little daunting.  After a little bit of introduction she jumped in with a number of more-or-less disturbing examples of robots that are out there now, some on the market, some in more of a research capacity.

For me, the most interesting, and most immediately pressing from an ethical point of view, is the robot teacher from Korea.  It's not fully autonomous - it has a human operator.  But the operator isn't controlling every minute motion of the robot.

The ethical questions she brought up around this particular example were on two lines:

1) the changing of the teacher-student relationship - normally it's dominant-submissive. But person-robot is as well, and when you make the teacher a robot, suddenly you have a dominant role but a submissive entity...

2) uncertainty around such young children being exposed to and interacting with a telepresent person - will they be able to tell the difference between that and a robot and a person?

Both are interesting questions.  My gut feeling is that the second is less of a problem than some people might think - kids are pretty good at sorting out complex rules of interaction with different people.  I suspect they'll handle this new telepresent person-entity just fine.

But there's definitely room for research into both - how do kids bond and interact with such artificial entities?  How do people behave when you have contradictory role and entity relationship modes?  Science and research can provide a lot of data to inform ethical decisions like that, and it's good to have people like AJung bringing them up for consideration.

I think I'll be going to more of these events.