Purpose This study aims to synthesize the “Sentient Infrastructure” paradigm, a transformative shift from reactive to proactive civil infrastructure management. The author critically analyze the synergistic integration of a sensor “Nervous System,” a dynamic “Digital Brain” (Digital Twins) and an artificial intelligence (AI) “Cognition Engine,” supported by a secure Internet of Things “Circulatory System.” By framing these technologies into a bio-inspired, closed-loop architecture, this work provides a holistic blueprint for creating intelligent, self-aware cyber-physical infrastructure capable of autonomous diagnosis, prognosis and adaptive resilience, ultimately paving the way for safer and more sustainable infrastructure management. Design/methodology/approach This review adopts a systematic, multidisciplinary approach integrating civil engineering, materials science and AI. It synthesizes over 100 recent studies on novel sensors, dynamic digital twins and AI-driven diagnostics to construct a unified *Sentient Infrastructure* framework. The methodology includes comparative analyses of sensing technologies, mathematical modeling of data assimilation using Ensemble Kalman Filters, and evaluation of AI models such as physics-informed and generative neural networks. Additionally, real-world case studies illustrate practical implementations and performance metrics, validating how the integration of sensing, modeling and cognition enables adaptive, real-time and resilient infrastructure management. Findings The study reveals that integrating advanced sensors, dynamic digital twins and AI-driven cognition transforms infrastructure from passive systems into adaptive, self-aware entities. Novel sensors like TENGs and distributed fiber optics enable dense, continuous monitoring; data assimilation through Ensemble Kalman Filters ensures real-time model fidelity; and physics-informed AI enhances diagnostic accuracy and remaining useful life prediction. Case studies confirm up to 80% improvement in model precision and 99% reduction in damage detection time. Collectively, these technologies establish a closed-loop framework that enables predictive maintenance, operational efficiency and resilient infrastructure management across diverse civil applications. Research limitations/implications While the Sentient Infrastructure paradigm shows transformative potential, its large-scale deployment faces several limitations. High implementation costs, lack of interoperability standards and uncertainties in long-term sensor durability pose practical challenges. Additionally, the complexity and opacity of AI models raise trust and explainability issues for safety-critical decisions. Data governance, privacy and cybersecurity concerns also remain unresolved. These limitations imply a strong need for interdisciplinary research in federated learning, neuromorphic edge computing and explainable AI to ensure scalable, transparent and secure adoption of sentient infrastructure systems in real-world civil engineering and smart city environments. Practical implications The Sentient Infrastructure paradigm offers significant practical benefits for engineers, asset managers and policymakers. By integrating self-powered sensors, dynamic digital twins and AI-driven analytics, it enables real-time condition assessment, predictive maintenance and optimized resource allocation. This reduces inspection costs, minimizes downtime and enhances public safety. Case studies demonstrate measurable improvements in model accuracy, early fault detection and lifecycle management. The framework also supports smart city development through system-of-systems integration, allowing coordinated management of bridges, tunnels and utilities. Overall, it provides a scalable pathway toward resilient, data-driven and self-adaptive civil infrastructure systems. Social implications The Sentient Infrastructure paradigm holds profound social benefits by enhancing the safety, reliability and sustainability of critical public assets. Continuous, intelligent monitoring reduces the risk of catastrophic failures, protecting lives and communities. Predictive maintenance optimizes public spending, freeing resources for social development. Moreover, resilient infrastructure strengthens disaster preparedness and climate adaptability, ensuring uninterrupted mobility and essential services. The integration of explainable AI fosters transparency and public trust in automated decision-making. Ultimately, this paradigm contributes to building smarter, safer and more sustainable cities that improve quality of life and societal resilience. Originality/value This paper introduces the *Sentient Infrastructure* paradigm as a novel, integrative framework that redefines civil infrastructure as a self-aware cyber-physical system. Unlike prior studies focusing separately on sensors, digital twins or AI, it unifies these domains into a cohesive architecture comprising a “nervous system,” “digital brain,” and “cognition engine.” The work’s originality lies in its systems-level synthesis, the introduction of bioinspired analogies for infrastructure intelligence and the use of physics-informed and generative AI for real-time diagnosis and prognosis. It provides a visionary yet practical roadmap for transforming infrastructure management from reactive to predictive and autonomous.
Belghachi Mohammed (Sat,) studied this question.