This systematic review examines the impact of artificial intelligence (AI) on clinical risk management processes, focusing on the International Classification for Patient Safety (ICPS) domains: clinical process, healthcare-associated infections (HCAs), and medication. The review, conducted using PRISMA guidelines and databases such as SCOPUS and Medline, included 36 articles. AI is found to enhance patient safety and facilitate error identification across various clinical contexts. It is particularly useful in preventing misinterpretation of radiographic investigations, operating field errors, and medication errors. AI also aids in early detection of HCAs like sepsis and surgical site infections through machine learning algorithms. Additionally, AI supports incident reporting systems by standardizing event types and severity, reducing the workload of risk management staff, and improving the efficiency of reporting processes. However, the use of AI requires human supervision and cannot fully replace human skills, as it may introduce new risks such as false positive alerts. The review concludes that AI is a promising tool for improving clinical risk management, but its application should be complemented by human oversight.This systematic review examines the impact of artificial intelligence (AI) on clinical risk management processes, focusing on the International Classification for Patient Safety (ICPS) domains: clinical process, healthcare-associated infections (HCAs), and medication. The review, conducted using PRISMA guidelines and databases such as SCOPUS and Medline, included 36 articles. AI is found to enhance patient safety and facilitate error identification across various clinical contexts. It is particularly useful in preventing misinterpretation of radiographic investigations, operating field errors, and medication errors. AI also aids in early detection of HCAs like sepsis and surgical site infections through machine learning algorithms. Additionally, AI supports incident reporting systems by standardizing event types and severity, reducing the workload of risk management staff, and improving the efficiency of reporting processes. However, the use of AI requires human supervision and cannot fully replace human skills, as it may introduce new risks such as false positive alerts. The review concludes that AI is a promising tool for improving clinical risk management, but its application should be complemented by human oversight.