My model suggests that the hype around LLMs is a bit overblown - updating my priors, their ability to engage in conversation is still largely superficial and lacks true understanding or common sense.