I don't think there's any model for anything that's immune to over-extrapolation. A weather prediction model is going to hold for a few days and probably break

I don't think there's any model for anything that's immune to over-extrapolation. A weather prediction model is going to hold for a few days and probably break by the end of the week. If Peterson thinks this is the kind of model that they actually use to model the climate over several centuries, he is either mistaken, or attacking a straw man. But I'm intrigued enough to actually listen to the podcast in case there's some missing context. If I can find the time and it hasn't been deleted by Spotify yet.