The construction industry has the highest number of fatal accidents compared to other industries. Manual monitoring of safety compliance is complex and challenging for safety engineers, so more automated solutions should be found. The research objectives were to explore the methods used for real time risk assessment in the construction sector, so a systematic review is proposed to summarise these new methods. Following the PRISMA methodology, three databases, Scopus, Web of Science and Inspec, were searched using four keywords: “construction”, “risk assessment”, “risk management” and “real time”. Only studies using real time risk assessment were selected, resulting in 18 articles. Several real time risk assessment technologies based on artificial intelligence were identified. These technologies can alert workers to a risky situation at any time, thus preventing an accident. This study showed that, in addition to more traditional risk assessment, new technologies for real time risk assessment are being introduced, changing the approach from reactive to proactive, allowing this monitoring to improve workplace surveillance and safety. Further studies should be developed to increase the sample studied, to increase confidence in the results and to assess the cost/benefit of implementation in a real-world context.