Stuart_Armstrong comments on When the uncertainty about the model is higher than the uncertainty in the model