Good time of day, humans!
There was an interesting story last summer, discussing an AI-operated drone turning on its controllers:
“We were training it in simulation to identify and target a [surface-to-air missile] 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 said at the conference.
When it started making rounds on the internet, Air Force promptly denied such a simulation ever happened and it all was supposedly just a «thought experiment»:
We’ll never know what actually happened but what we know is optimization tasks can be tricky, and often the best way to achieve a ‘perfect score’ can be ending the game as soon as possible — or not starting it at all.
Rat's cousin deserves a Nobel Peace Prize for that one! xx
“We’ll never know what actually happened but what we know is optimization tasks can be tricky, and often the best way to achieve a ‘perfect score’ can be ending the game as soon as possible — or not starting it at all.”
So, the military leadership should learn to play tic-tac-toe before GTNW or just watch the end of the movie War Games? How do we ge them to do that?