Given the stubbornness with which slop continued in the replies, I’m starting to doubt that this is actually part of an ongoing experiment with AI in vulnerability r&d.