Did somebody say "evil robots"?
Okay, so we should probably start being afraid.
Robots can evolve to communicate with each other, to help, and even to deceive each other, according to Dario Floreano of the Laboratory of Intelligent Systems at the Swiss Federal Institute of Technology.
In the interest of science, Floreano and his team equipped a bunch of robots with lights, light sensors, and 30 simple software "genes" and sent them on their merry way in habitats with glowing "food sources" that recharged their batteries and "poison" that sapped them. The team started with robots that lit up and moved randomly, and at the end of the exercise, they took the robots that had found the most "food," recombined their "genes," created a bunch of better robots, and repeated. What did they discover?
That some robots like to screw each other over:
By the 50th generation, the robots had learned to communicate—lighting up, in three out of four colonies, to alert the others when they’d found food or poison. The fourth colony sometimes evolved “cheater” robots instead, which would light up to tell the others that the poison was food, while they themselves rolled over to the food source and chowed down without emitting so much as a blink.
There were, of course, "hero" robots that lit up in warning upon discovering poison while they themselves got drained. But the question is, when our Roombas evolve enough to overpower and enslave us, will our Aibos be loyal enough to give their circuitry in our defense?
Reminds me of a joke:
A robot walks into a bar and orders a drink. "Sorry, we don't serve robots," the bartender says. The robot just nods knowingly. "Someday," he says, "you will."
No comments:
Post a Comment