I got an email asking what's HN's policy on GPT-3 generated comments. I think it's covered by the fact that we don't allow bots on HN (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...). The purpose of the the threads is human conversation.
Obviously there's an infatuation right now with GPT-3. That's normal. If people keep posting these without disclosing them, I imagine there will be two consequences. One (good for HN) is readers scrutinizing comments more closely and raising the bar for what counts as a high-quality comment. The other (bad for HN) is readers accusing each other of posting generated comments.
Accusing other commenters of being bots is not a new phenomenon ("this sounds like it was written by a Markov chain" has long been an internet swipe) but if it gets bigger, we might have to figure something out. But first we should wait for the original wave of novelty to die down.
Obviously there's an infatuation right now with GPT-3. That's normal. If people keep posting these without disclosing them, I imagine there will be two consequences. One (good for HN) is readers scrutinizing comments more closely and raising the bar for what counts as a high-quality comment. The other (bad for HN) is readers accusing each other of posting generated comments.
Accusing other commenters of being bots is not a new phenomenon ("this sounds like it was written by a Markov chain" has long been an internet swipe) but if it gets bigger, we might have to figure something out. But first we should wait for the original wave of novelty to die down.