Large Language Models (LLMs) can generate empathic responses perceived as more empathic than human-written responses, showcasing potential for enhancing human peer support.
