He seems completely dismissive, and rather certain that GPT doesn't have a 'world model', despite IMHO not really having enough of an understanding (since no-on

He seems completely dismissive, and rather certain that GPT doesn't have a 'world model', despite IMHO not really having enough of an understanding (since no-one does, really), to justify that certainty. GPT has been trained on text, and so any 'world model' is going to be a step removed from the world, compared to systems that have been trained on real-world stimulii (vision, sound, touch) like animals and humans. But that does not mean it doesn't have a 'world model' at all, just that it's models are not going to be something we immediately recognize.