<p>Miners’ safety behavior is a critical factor in mine safety management, and its implementation efficacy considerably impacts the establishment of safety ecosystems as well as sustainable industry development. To overcome this problem, mining industries use a Wireless Sensor Network (WSN). Even though WSN decreases the risk factors, there are some issues, like radio connectivity and the complex nature of wave propagation in the mining industry. Autonomous technologies like Artificial Intelligence (AI) and Machine Learning (ML) [Subset of AI] models are incorporated with the mining industry to tackle the concerns by enabling more accurate hazard detection and predictive risk assessment in the Underground Mines (UMs). Therefore, in this review, the main safety hazards in modern mining operations are explored via a systematic review kind of approach and the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) framework. Also, AI- and ML-based models have been applied to address the importance of objective, highlighting outcomes like improved detection accuracy and improved worker protection. Additionally, the methodological gaps, particularly in data quality, model generalization, and real-time deployment, are explored in this review. This review finds future research directions aimed at integrating robust AI-driven risk assessment frameworks into mining safety management systems.</p>