April 26, 2012

Bad Robot or Bad Programmer? Who is Accountable?

I love reading about research done at various different universities around the world. I'll admit that I love the stories even more when the research amuses me in some shape or form.

Though if I'm perfectly honest the latest research from the University of Washington in Seattle makes me feel a bit uneasy. You see, it's about how humans and robots interact.

Here's a bit of background for you.

Forty students took part in a scavenger hunt where they had to find a given amount of items within a certain time, then a robot named Robovie judged whether or not they should win the $20 prize money.

The only problem with this was, Robovie was programmed to lie.

Now, the strange thing is, even though the students saw Robovie as a robot and didn't consider it to be morally accountable for anything, some still got really annoyed when it lied. Realistically their anger should've been with the programmer rather than the inanimate object doing what it was 'told' to do.

Heather Gary, a doctoral student in developmental psychology at the University of Washington in Seattle said, "Some accused Robovie of lying or cheating."

About 65 percent of the 40 students said Robovie was at least somewhat morally accountable for lying.

Just as I was considering the possibility, with some degree of smugness, that I would realise that old Robovie was mere nuts, bolts and electrical wiring I thought of something else.

I have a name for my car & love my iPad, I shout at my work's PC when it locks up and freezes. Why do I do that when I know that whatever emotion I have for a piece of machinery won't change it's behaviour?

Yup, I hold machinery morally accountable for things, I give my car steering wheel a pat if it's gotten me home with hardly any fuel in it or it's started first time when the outside temperature is -21C.

Based on the findings in their paper 'Do Humans Hold a Humanoid Robot Morally Accountable for the Harm It Causes?' Heather Gary and the rest of her colleagues suggested that the military should consider the moral accountability of robots. I'll take that one step further and suggest that everyone or company that are building humanoid robots and the like should consider her findings.

B1-66ER robot that didnt want to die
It all reminds me a bit of 'Animatrix' the short stories based on and around the film trilogy 'The Matrix'. Ok, ok, I know that those stories are based on robots with artificial intelligence, but is that really so far away in the future?

After all....... It was only a year ago that Watson the IBM computer appeared on the tv programme 'Jeopardy'