AI-enabled flexible sensing technologies for safety engineering: Materials, devices, and applications

Safety engineering underpins the resilience and stability of modern industrial and societal infrastructures; however, conventional rigid monitoring systems remain constrained by limited conformability, adaptability and real-time responsiveness in complex and dynamic environments. The convergence of flexible electronics and artificial intelligence (AI) offers a transformative route from passive signal acquisition to active cognitive decision-making. Here, we provide a systematic overview of AI-enabled flexible sensing systems for critical safety applications. We first categorize functional materials by dimensionality and discuss material design strategies that enable mechanical compliance, environmental robustness and sensing capability. Building upon this materials foundation, we further examine computational architectures that transform multimodal sensing signals into actionable intelligence, with emphasis on deep learning-based diagnostics, edge computing frameworks, and neuromorphic devices that alleviate centralized processing constraints. We contextualize these advances across representative safety scenarios, including structural health monitoring, occupational risk assessment, power infrastructure inspection, and fire detection. Finally, we discuss the key challenges, including reliability, scalability, and long-term stability, that must be addressed to enable the transition from assisted monitoring toward autonomous and resilient safety systems. By bridging materials innovation and intelligent computation, AI-enabled flexible sensing technologies are positioned to shift safety engineering from reactive protection to predictive and cognitive paradigms.