If a GPT bot can even fool /pol/ 4channers, then Reddit is just like easy mode, as I think Reddit community is much more gullible (no offence, I visit Reddit a lot and recently also read on some Lemmy instances as well). That being said, it is interesting that people are starting to be more aware of "unnatural" answer and can point out the AI-generated content, like those GPT-3.5 responses in the beginning of the article.