Comment by amelius

Fun fact ... if we continue training LLMs on human data, in the future when you ask Claude a question it will say: "Have you asked Claude?"