With the rapid advancement of industrialization and urbanization, emerging pollutants has brought unprecedented challenges to environmental protection and posed significant threats to human health. In this context, artificial intelligence (AI), leveraging its efficiency and precision, is gradually becoming a critical tool for emerging pollutant governance. This study reviews the current status and major challenges regarding emerging pollutant governance, and proposes an AI-based framework for managing emerging pollutants. In the screening phase, deep learning and natural language processing technologies are utilized to identify potential emerging pollutants from vast amounts of data, enhancing screening speed and accuracy. In risk assessment, machine learning models integrate multidimensional data to construct a dynamic evaluation system that can quantitatively assess environmental behaviors and health risks of pollutants in real time. In the control phase, AI technology enables intelligent monitoring, optimal technology selection, and dynamic regulation, promoting continuous optimization of governance strategies. Furthermore, the study proposes a large model framework for emerging pollutants, aiming to integrate multimodal environmental data to assist in the identification, risk assessment, and optimization of governance strategies for emerging pollutants. Research recommendations include establishing an intelligent identification and monitoring system for emerging pollutants, developing a data-driven risk assessment and prediction platform, optimizing pollution control technology and management platforms, and building a knowledge-driven large-model-assisted decision-making system. These efforts aim to precisely improve AI-based governance of emerging pollutants, providing references for scientific research, industry applications, and policy-making in related fields.