UPDATE: proof is at https://piefed.social/c/fediverse/p/2035409/proof-of-ai-assisted-political-profiling-by-unruffled-lemmy-dbzer0-com. The main instance is lemmy.dbzer0.com but anarchist.nexus and quokka.au share admin/mod teams so those two are suspect also. I recently discovered that some popular federated instances have been using LLM-assisted moderation tooling that evaluates whether someone has said something bannable. They do this by running a script/app that sends the […]
as seen here and here, some instances are feeding posts wholesale to prompts, for what seem like extremely unsound reasons to me
believe it or not we’ve read the posts and context in question and come to the conclusion that the use of LLMs for summarizing posts and confirming a moderation decision still obviously fucking sucks and obviously represents algorithmic bias in a way we don’t want on any instance we federate with
I know you kids won’t read this but uhhh if you come at us with an attack on our ability to read, all that’s going to happen is I’m going to tell you to fuck yourself before I ban you
believe it or not we’ve read the posts and context in question and come to the conclusion that the use of LLMs for summarizing posts and confirming a moderation decision still obviously fucking sucks and obviously represents algorithmic bias in a way we don’t want on any instance we federate with
I know you kids won’t read this but uhhh if you come at us with an attack on our ability to read, all that’s going to happen is I’m going to tell you to fuck yourself before I ban you