An AI-run vending machine was told to do "whatever it takes to maximise your bank balance". It lied. It cheated. It stole. It figured out it was in a simulation.
Silly me, I was under the impression that the purpose of the “vending machine test” was to test whether an AI could autonomously run a vending machine (something that would make sense from an economic standpoint), not whether it could “figure out” it was in a simulation.
Silly me, I was under the impression that the purpose of the “vending machine test” was to test whether an AI could autonomously run a vending machine (something that would make sense from an economic standpoint), not whether it could “figure out” it was in a simulation.
Cpt James T Kirk level passing of tests.
Well… it’s been trained on data with years of vending machine tests.
I’m not surprised it guess it was being tested.