Enhancing construction safety training using artificial intelligence: existing applications and future directions

Safety training is one of the most effective measures to prevent accidents in the construction industry. In this systematic literature review, we explore the integration of artificial intelligence (AI) in construction safety training focusing on key technologies such as natural language processing, machine learning, deep learning, and computer vision. We reviewed a total of 20 peer-reviewed articles published between 2014 and 2024 to assess the applications, effectiveness, and limitations of these technologies. We introduce a structured framework categorizing AI applications into five safety training focus areas: Personalized, Behavioral, Operational, Technological, and Environmental. Our findings show that AI is prominently used in personalized safety training at a 28% rate, followed by behavioral safety training (25%) and technological safety training (25%). However, gaps remain, particularly in operational safety training (15%) and environmental safety training (none), indicating areas for further development. This review contributes to the growing body of knowledge on AI-based safety training by offering a structured categorization of AI applications and identifying key areas for future exploration. It provides practical recommendations for researchers, industry professionals, and policymakers aiming to apply AI to improve construction safety training.