In recent years, the escalating risks of healthcare-associated infection (HAI) and the transmission of multidrug-resistant organisms have emerged as significant global public health challenges, posing a grave threat to medical care quality and safety. HAI prevention and control are confronted with issues such as pathogen transmission complex routes, dynamic changes in infection risks of specific populations, and the lag in traditional monitoring methods. Traditional HAI management model relies on manual monitoring and information systems, presenting predicaments such as low efficiency, fragmented data, and delayed warnings. Artificial intelligence (AI) technology integrates electronic health records (EHRs), vital signs, and other clinical data to develop predictive models based on machine learning (ML) and deep learning (DL), and has enhanced multimodal data fusion and real-time dynamic analysis capabilities, demonstrating significant advantages in risk prediction, early diagnosis, and precision intervention of HAI. This paper systematically reviews the developmental trajectory, scientific achievements, and innovative practices of AI technology in HAI management, delves into existing bottlenecks such as data quality, algorithm relia-bility, and ethical norms, aiming to provide theoretical and practical references for establishing intelligent and precise HAI prevention and control system.