Abstract: The construction of intelligent mines is driving the transformation of coal mine disaster prevention and control toward intelligent development, under which the prevention and control of coal and gas outburst disasters has become a critical challenge that urgently needs to be addressed in deep mining conditions. However, the existing coal and gas outburst prevention and control system still has problems such as insufficient system integrity, broken prevention and control chains, and limited level of intelligence, making it difficult to achieve advanced and precise risk prevention and control. This article aims to construct a theoretical framework for the intelligent dual prevention mechanism of coal and gas outbursts, clarify its research boundaries, systematically review its research progress, thereby providing theoretical support for the establishment of an intelligent coal and gas outburst prevention and control system. This article is based on the theory of dual prevention mechanism, taking coal and gas outburst disasters as the research object. Using a combination of literature analysis and systematic induction, it systematically sorts out and summarizes the intelligent dual prevention mechanism for coal and gas outbursts from five aspects: risk identification, assessment, prediction, warning and control. It summarizes the current research status, contributions and problems, and proposes future research priorities and directions. The results indicate that: The intelligent dual prevention mechanism for coal and gas outbursts covers risk identification, assessment, prediction, warning and control, as well as hazard classification and grading, investigation, treatment and acceptance, forming a systematic framework for disaster prevention and control. In terms of research contributions, intelligent identification has achieved automatic extraction of unstructured data and multi factor coupling analysis, intelligent assessment has constructed a dual driving paradigm of multi-attribute decision-making and machine learning fusion, intelligent prediction has promoted the evolution from single point perception to multi-source fusion, intelligent warning has established a three-layer system of “real-time assessment-advanced prediction-precursor monitoring”, and intelligent control has promoted the development of system integration, collaborative intelligence, and closed-loop control. In terms of research issues, the current intelligent identification system is not yet mature, the intelligent assessment model is fragmented and lacks generalization ability, the intelligent prediction has shallow data fusion and real-time deployment difficulties, the systematic exploration of intelligent warning is insufficient, the closed-loop efficiency of intelligent control is limited, resulting in weak coordination of the entire chain for prevention and control. Future research should establish an intelligent dual prevention system for coal and gas outbursts, deepen the fusion of multi-source data, qualitative quantitative transformation of indicators, and unified real-time and periodic risks in identification, develop multi-level fusion frameworks and small sample, transfer learning methods in assessment, break through the difficulties of mechanism data fusion and model deployment in prediction, improve dynamic warning rules, multi device linkage and control response mechanisms in early warning, establish adaptive, collaborative intelligence and closed-loop mechanisms for control, and ultimately achieve full chain closed-loop intelligent prevention and control of coal and gas outburst disasters.