<p>Antimicrobial peptides (AMPs) have emerged as promising alternatives to traditional antibiotics for combating antimicrobial resistance, owing to their unique mechanisms of action and low propensity for resistance development. As antibiotic resistance escalates, there is an urgent need for novel antimicrobial strategies. Artificial intelligence (AI) technologies, particularly machine learning (ML) and deep learning (DL), now offer unprecedented opportunities for accelerating AMP discovery and design. Current AI applications span discriminative models, regression models, generative models, and multimodal optimization, significantly improving screening efficiency, enabling innovative design strategies, and facilitating pre-clinical validation. However, AI-driven AMP research still faces challenges including data quality limitations, model interpretability, and experimental validation bottlenecks. This review systematically summarizes the latest AI advances in AMP research, analyzes key technical hurdles, and outlines future directions and emerging opportunities, providing researchers with comprehensive theoretical and practical guidance to expedite AMP-based drug development.</p>