Jake Mendel on X: "@herbiebradley @MaxNadeau_ @theojaffee Max is referring to conceptual arguments about challenges with avoiding cheating and lying that will arise in the future. I think the claim that for sufficiently capable AI systems it becomes very difficult to design RL environments that don't incentivize cheating is pretty hard" / X
x.com · saved by 1 readers
Max is referring to conceptual arguments about challenges with avoiding cheating and lying that will arise in the future. I think the claim that for sufficiently capable AI systems it becomes very difficult to design RL environments that don't incentivize cheating is pretty hard to contest, and cur…