The good old AI question. When AI becomes so good that you can't detect it any longer, what will we be upvoting then? Just people? Their vision? Their opinions? Their contributions? Their goals? Their promises?
I think it's safe to say content hasn't been the main priority of curation for a long time now, and that's fine in many different ways. The main issue I see with the rise of AI is trust in newcomers and farms which may or may not be happening already while some users may already be relying on AI generating most of the content and even providing ideas on what to post today for your automatic upvotes. This is not one of those posts, however.
This discussion stemmed from me asking if an AI account greeting newcomers is a good thing when someone was constantly autovoting its comments to the top past all the genuine comments. My thought process was that if I was a newcomer who was tired of seeing AI everywhere in their current social media platforms who finally managed to get an account and start posting here, their first impression may not be the best to first see an AI generated comment greet them. I'm not saying the AI comments/bots are bad per se, but I don't think they should be placed above regular users as there's still quite a few regular users welcoming newcomers which is nice to see.
Another thing that happened recently was a new project that had posted in a community I curate in and support, their posts were AI generated and they were accepting rewards which I think is safe to say is quite frowned upon generally on Hive. The community account had muted the posts first but then had noticed that the project owner had been selfvoting the posts as well with their other project, which had resulted in the community account also casting a few downvotes on the posts. The project owner, furious about what had occurred complained that it didn't state in the community rules that AI generated posts aren't welcomed and that he may do as he please with his delegated voting power, to which he then had gone and downvoted several comments of the community account and shockingly even taken it so far to target innocent users posting into the community which I felt was a bit too far.
Either way, there will be a time where we'll have agents that have downloaded say all the posts you've written on Hive and have learned to write like you and about things you write about to be able to continue doing so. I can imagine I could probably create such an agent right now if I wanted to, it'd probably be quite a weird one trying really hard to create super long and confusing sentences trying to mimic me. My morals aren't letting me go down that route however, plus if caught I'd fully expect the community to turn against me and downvote the posts which I'd personally agree to as things currently stand. Would you?
But the bigger question is how can we know when that happens for others? How do we know if newcomers are writing their posts or if it's all just AI? How do we know if they're genuine users or just one bot operator running hundreds of accounts pretending to be different people?
I don't have a good answer for that right now, but it is things we're going to need to tackle at some point.
Focusing curation on users we trust is probably going to be a big thing in the near future, which may not be great for the barrier of entry that I've posted about, people being able to easily come here and quickly be able to earn a few coins compared to other places if we're full of distrust of any and all newcomers being AI. This also begs to question distribution of stake and centralization of inflation, but I guess that's not that big of an issue given how much hive exists on exchanges these days.
Anyway, kind of a dystopian odd post for today I guess with not much reassurance or providing any solutions, but some food for thought I guess.
[img created with grok]
