Sparks of Artificial General Recommender (AGR): Experiments with ChatGPT

This study investigates the feasibility of developing an Artificial General Recommender (AGR), facilitated by recent advancements in Large Language Models (LLMs). An AGR comprises both conversationality and universality to engage natural dialogues generate recommendations across various domains. We propose ten fundamental principles that should adhere to, each with its corresponding testing protocol. proceed assess whether ChatGPT, a sophisticated LLM, can comply proposed engaging recommendation-oriented model while observing behavior. Our findings demonstrate potential for ChatGPT serve as AGR, though several limitations areas improvement are identified.