It’s hard to say. This form of text post LLM farming is relatively new.

Things that are older include bots posting old popular posts and comments from different subreddits. That’s harder to catch just with normal browsing.

Then there are various changes that have seemed to lower the quality of posts over the last several years, like only posting screenshots of tweets, or screenshots of headlines/blurbs (i.e. removal of the source, with nobody caring to post it in the comments; for the hundreds of comments, the headline or blurb is apparently, without fail, all they need), and a strange tick-up in non-native english in post titles that could be outsourced human sock puppet farming (because it often can be seen paired with US politics posts), or just reddit getting more popular in India. Basically, is all of that a sign of bots/sock puppets, or is it just the vast majority of internet users now not knowing how to do much more on a device than screenshot and not caring to know much more than three sentences about any given subject or event? Are we living in a kind of early-cyberpunk 1984, or Wall-E?

But one scary thing is that you can of course imagine the strategy seen in the picture, but ‘good’, i.e. much harder to track because somebody put a little more effort into variety of title, content and style. And there’s no reason why you couldn’t use an AI to post things that look more like the traditional reddit post.

And even this proof of concept level attempt has thousands of upvotes each and what look like real users engaging with the content, with no user, moderator, or admin seeming to particularly notice or care.