Jeremy Shere

Scientists and engineers at MIT have created a new, experimental robot called Nexi that can appear to express human emotions like anger, surprise and sadness through a series of facial expressions.

The machine is an MDS robot, which stands for mobile, dexterous and social. And, indeed, as you can see in this clip, Nexi is all of those things …

Which raises a few questions. First, why do we want robots to act and appear as human as possible? Do we? I few years ago I remember reading an article in Popular Science about a roboticist named David Hanson who made a working robot head that looked exactly like his girlfriend. Using several motors and other electronic doo-dads covered by a flexible plastic skin molded to look like his girlfriend, Hanson could make the head move and model various facial expressions. And it freaked people out because of a phenomenon called the Uncanny Valley. The basic idea is that people are attracted to robots and avatars that look human–up to a point. But beyond that point–when the robot looks almost but not quite human–people are repulsed. An article in the online magazine Slate makes the point that however good the graphics are in video games, renderings of human figure always come off as decidedly creepy when you see their faces close up. The bodies seem alive but the faces look dead, devoid of any human spark.

Not that this is a problem with Nexi, which looks nothing like a real person and very much like a robot. But Nexi is clearly a step in the direction of making robots look and act like actual people. And I’ve always wondered–why do we feel the need or desire to recreate ourselves in robot form? The most useful robots are ones that look like machines–robot arms on assembly lines, those robot vacuum cleaners that scoot around the floor like oversize hockey pucks … while human-like robots are essentially useless, as far as I can tell. Even in Star Wars, C-3PO was always standing around annoying everyone while R2-D2 did all the hard work. (Although I suppose Data from Star Trek Next Generation was pretty cool, to be fair.)

I’m not trying to make a utilitarian argument. Research and invention are not (nor should they be) necessarily spurred by function or need. Creativity, curiosity and invention are just as important. But this desire for robots that look like us still strikes me as curious. In movies and science fiction literature, at least, androids are almost always dangerous and feared–I, Robot, Aliens and its sequels, 2001, Blade Runner, etc.

Anyhow, I wonder what you think. Would you look forward to a future where robots have become so much like us that it’s hard to tell the difference?