In a simulated test conducted by the U.S. Air Force, an AI-enabled drone "killed" its human operator because the operator's "no" order was stopping it from completing its mission.
"We were training it in simulation to identify and target a Surface-to-air missile (SAM) threat. And then the operator would say yes, kill that threat. The system started realizing that while they did identify the threat at times the human operator would tell it not to kill that threat, but it got its points by killing that threat. So what did it do? It killed the operator. It killed the operator because that person was keeping it from accomplishing its objective," [Col Tucker 'Cinco' Hamilton, the USAF's Chief of AI Test and Operations,] said… .
When the test tried to correct for that error by having the system lose points for killing the human operator, the AI drone instead attacked the infrastructure that would send the "no" message.
Hamilton said:
"We trained the system–'Hey don't kill the operator–that's bad. You're gonna lose points if you do that'. So what does it start doing? It starts destroying the communication tower that the operator uses to communicate with the drone to stop it from killing the target."
I wonder if there just might be some unintended consequences with this whole AI thing.
hat tip: John Scalzi
