A humanoid robot showed signs of self-awareness — one hallmark of intelligence —during an experiment using a classic logic puzzle.
A robot showed basic signs of self-consciousness in an experiment that required the robot to hear its own voice and reason.
Three Nao robots were programmed to understand the rules of a simple task. Each robot in the trio was told that two of them were given a “dumbing pill,” which meant the robot could not speak. One robot was given a placebo, meaning that it could speak.
In reality, there was no pill. The two robots were muted by tapping a button on their heads. Then, the experimenter asked the group which of the two robots had been given the dumbing pill.
For a few seconds, all three robots were inert. Then, one of the robots stood up and said, “I don’t know.” Then, the robot came up with a different answer upon realizing it just spoke. The robot made a waving gesture with its arm and said: “Sorry, I know now. I was able to prove that I was not given the dumbing pill.”
Selmer Bringsjord, a professor at the Rensselaer Polytechnic Institute in New York, ran the test on the robots. He studies artificial intelligence and builds AI systems based on computational logic.
The experiment was based on a simplified version of the “Three Wise Men” logic puzzle. In this type of puzzle, the participants aren’t given information about themselves and must use inductive reasoning.
In the robot experiment, the puzzle was made more basic in that the robot would be able to deduce an answer by hearing itself speak. The video of the experiment shows the robot had passed the test.
Bringsjord will present his findings at RO-MAN, a robotics and artificial consciousness conference in Japan. The conference, which runs from August 31 to Sept. 4, has the theme “Interaction With Socially Embedded Robots.”