1 paper
Rudolf Laine, Bilal Chughtai, Jan Betley +6
AI assistants such as ChatGPT are trained to respond to users by saying, "I am a large language model". This raises questions. Do such models know that they are LLMs and reliably a…