The masonry sector remains one of the most labor-intensive and accident-prone areas of construction, where strenuous manual work and complex site conditions continue to drive high injury rates despite established safety measures. This review examines the state, challenges, and future directions for large-scale adoption of Artificial Intelligence (AI) and sensing technologies for enhancing masonry safety within the emerging Industry 5.0 paradigm. Using a three-stage methodology, the study first applied the PRISMA protocol to screen the Scopus database, identifying 128 relevant articles. The selected works were systematically categorized into four domains – safety application areas, dataset development, AI methods, and sensing technologies – followed by comparative evaluation of their performance, advantages, and limitations. A critical synthesis then identified cross-cutting barriers and research gaps. Results reveal that Computer Vision (CV) and Natural Language Processing (NLP) enable real-time hazard detection, accident classification, and predictive analytics while wearable sensors and Unmanned Aerial Vehicles (UAVs) support physiological and environmental monitoring, achieving accuracies above 80-90% in controlled settings. However, limited masonry-specific datasets, environmental variability, sensing errors, hardware costs, and privacy concerns hinder scalability. The review concludes that progress depends on standardized datasets, algorithms resilient to field conditions, and integrated multimodal AI-driven sensing systems aligned with human-centric Industry 5.0 principles. Emerging technologies such as collaborative robotics and Virtual Reality (VR) also hold promise for advancing masonry safety. The insights presented provide a roadmap for researchers, practitioners, and policymakers toward scalable, ethical, and proactive safety innovation. 

Masonry
Safety 
Systematic review 
AI
Computer vision
Sensors