“An” issue with it is that we may come to rely on these AI’s outputs as assumed correctness or truth. If we have to double check everything they produce then th

“An” issue with it is that we may come to rely on these AI’s outputs as assumed correctness or truth. If we have to double check everything they produce then that’s not great either.