Iv been pondering this a bit recently. At some pt if the models are smarter than us are we going to be smart enough to tell the difference? (assuming they remain sycophantic to people which is a big if)