A robot allegedly demonstrates basic self-awareness while attempting to solve a classic logic puzzle.
Researchers at the Rensselaer Polytechnic Institute AI and Reasoning Lab in New York adapted the classic inductive reasoning puzzle known as The King’s Wise Men and posed the problem to a trio of robots.
In the puzzle, three wise men are given hats of either white or blue, with the guarantee that at least one of the hats is blue. The men are not allowed to speak with each other; the first man to stand up and correctly announce the colour of his hat wins.
For the robot’s self-awareness test, Selmer Bringsjord, chair of the department of cognitive science at the institute, used three of French robotics company Aldebaran’s humanoid Nao robots. He had programmed these with a proprietary algorithm called Deontic Cognitive Event Calculus, which enables the robots to carry out reasoning.
The three robots were told that two of them had been given “dumbing pills” that rendered them unable to speak, and one a placebo (in reality, they had a button pressed on their heads, but the result was the same). They were then asked if they had been given the dumbing pill or the placebo.
In the video below, which was posted earlier this month, you can see the results. There are several long moments of silence before one robot stands up and says “I don’t know.”
It then raises its hand like a child in a schoolroom, and offers a correction: “Sorry, I know now. I was able to prove that I was not given the dumbing pill.”
In order to demonstrate this sort of self awareness, the robot must be able to understand the rules of the puzzle, recognise its own voice and recognise that it is an individual distinct from the other two robots.
Read the full article.