Cynically this is a great way for the unhinged admin to train and harden an AI censorship filter :)
replyIf you’re in need of a writing prompt …
Nothing I can do will go as hard as a frontier model with zero safeties. Trust me. They are downright scary in how sneaky they get defeating classifiers and safeties.
reply