I begin to wonder how LLMS can be trained. They refuse to admit a lack of knowledge and when they are beyond the edge of their core training, they default to making things up.