I think that “imitate a human who is trying to be helpful” is better than “imitate a human who is writing an article on the internet,” even though it’s hard to define “helpful.” I agree that’s not completely obvious for a bunch of reasons.
(GPT-3 is better if your goal is in fact to predict text that people write on the internet, but that’s a minority of API applications.)
I think that “imitate a human who is trying to be helpful” is better than “imitate a human who is writing an article on the internet,” even though it’s hard to define “helpful.” I agree that’s not completely obvious for a bunch of reasons.
(GPT-3 is better if your goal is in fact to predict text that people write on the internet, but that’s a minority of API applications.)