Two models of AI oversight – and how things could go deeply wrong(garymarcus.substack.com)1 points by son_of_gloin 3 years ago | 0 commentsNo comments yet