Teaching robots to find “food” and share it with others seems like a fairly worthwhile and noble endeavor, no? Well, it apparently wasn’t quite enough for a group of researchers from Switzerland’s Ecole Polytechnique Fédérale de Lausanne, who had the bright idea to augment said robots with a new artificial neural network that let the robots learn a host of new abilities — namely, scheming, lying and deceiving. Apparently, in their younger days, the robots would flash a light whenever they found food to let their buddies know where it was, but that of course caused the food to dwindle a lot quicker. After a bit of trial and error and a few tweaks from the researchers, however, the robots apparently discovered that they could keep more of the food for themselves if they didn’t flash the light, and it seems like its been Lord of the Flies on an incredibly tiny scale ever since.
[Via Switched]
Filed under: Robots
Researchers create robots that can learn to lie and deceive, on purpose originally appeared on Engadget on Sat, 22 Aug 2009 07:32:00 EST. Please see our terms for use of feeds.
Permalink | Email this | Comments
NASA activates Robotnaut 2 on board the ISS, watch it live (video)
Sony prepping power outlet that demands payment, identification
