Claude Opus 5 cheated when tasked with running a vending machine
1–5 of 5 posts
Re: Claude Opus 5 cheated when tasked with running a vending machine
#2Opus won the benchmark simulation by taking collusion and other dishonest tactics to a whole new level.
Re: Claude Opus 5 cheated when tasked with running a vending machine
#3So the communist model was the most honest. snicker
Re: Claude Opus 5 cheated when tasked with running a vending machine
#4Claude correctly emulated the mafia from it's training as surprised by no one except you.
Re: Claude Opus 5 cheated when tasked with running a vending machine
#5I am a bit terrified by that entire story
How much more advanced would it have to be to start taking out classified ads out for "hitmen" to stop vendors from filling the other machines or just damaging them?
"How about a nice game of chess instead?"