How Poems Can Fool AI Into Breaching Its Own Safety Rules, And Help Create Nuclear Weapons Large language models (LLMs) are trained on staggering amounts of text, sourcing from books, research papers, news articles, code, conversations, and more.
