Robots Taught To Lie

Author Profile Picture
Entertainment Editor
09.13.10

Ronald Arkin and Alan Wagner at the Georgia Institute of Technology are teaching robots how to use deception. STOP DOING THAT.  They developed algorithms based on game theory and interdependence theory that helped a robot decide if it should use deception and how to go about deceiving a person or robot during a hide and seek exercise.  The sneaky hiding robot was able to trick the seeking robot 75% of the time by knocking over markers along one path, then hiding someplace else.

Robots on the battlefield, for instance, could use deception to elude captors. In a search and rescue scenario, a robot might have to be deceptive to handle a panicking human. For now, however, the robots are using their new skill to play a mean game of hide-and-seek. [Gizmag]

That’s what we need when we battle robots: ones better able to elude their captors (us).  And why not teach them how to take advantage of a panicking human as well?  SERIOUSLY, STOP DOING THAT.  Robots are already clanking contraptions of pure evil.  They don’t need help in being better con artists as well. Wait, a robot just knocked on the door. He says I look like I’ve lost weight and if I help him find his lost corgi puppy I can have some free candy and corgi cuddles. Awww, I take it back. Robots are harmless, delightful scamps.

Author Profile Picture
When not writing for Uproxx, Caleb likes to volunteer at the legless cat shelter and photoshop the Babadook into all of his family photos. He once resolved the question “To be or not to be?” through the clever use of General Semantics. Your mom thinks you could be more like him if you only applied yourself.

Around The Web