"thinking" and reasoning can be done by toddlers with a dataset a fraction of a fraction of the size that even the simplest language models are trained on. I d

"thinking" and reasoning can be done by toddlers with a dataset a fraction of a fraction of the size that even the simplest language models are trained on. I don't understand this thinking that it's x because it looks like x(thinking, artistic creativity, etc.). I can prompt Google for incrementally more correct answers to a problem, does that mean there's no difference between "google" and "thought"?