Thank you for raising this critical point, Dr. Božić. The application of AI in integrated risk management—especially in cybersecurity—is no longer optional; it’s becoming essential.

AI can support proactive risk identification, anomaly detection, and even dynamic threat modeling by analyzing patterns across vast and complex data environments. Machine learning models can help flag vulnerabilities, assess evolving threat vectors, and simulate risk scenarios much faster than traditional methods.

However, there are some caveats. Many AI systems function as black boxes, which complicates explainability—particularly in regulated environments like healthcare. Bias in training data, adversarial attacks on models, and over-reliance on automation can also introduce new layers of risk.

A responsible approach would involve integrating AI with human oversight, using explainable AI (XAI) techniques, and aligning risk models with frameworks like NIST RMF or ISO/IEC 27005. Done right, AI doesn’t replace traditional risk management—it enhances it by making it more adaptive and real-time.