You can tell a lot about a user based on their posting history, and GPT4 can pretty easily tell if that person has a history of good or bad faith interactions. I wrote a bot that uses GPT4 to do this analysis automatically, and while I think it would be very useful, and also potentially change the dynamics of online discourse, It feels scary and dystopian as well. I'm very torn on whether this sort of thing will be good or bad, but it seems obvious that this kind of thing will become ubiquitous. I took my bot down for now, but I'd love to hear arguments for and against it. What do you think?