Mitigating DNS Cybersecurity Risks through Coaching by Agentic AI: An Adaptive Network Analysis

This paper explores the human factors in threat detection and mitigation within the Domain Name System (DNS), focusing on the integration of an agentic AI as coach designed to support the decision making process of human administrators. A temporal-causal network model was developed to simulate a successful cache poisoning attack, incorporating both a mental model of an administrator's decision making process and a corresponding model of an agentic AI coach that detects misjudgements and provides corrective influence. A series of simulations were conducted to examine key variables such as fatigue and the administrator's susceptibility to its build-up, measuring the resulting latency in responsive actions. Results indicate that as the administrator's performance deteriorates under increasing fatigue, the agentic AI coach plays a critical role by initiating corrective steps at the knowledge level. The positive reinforcements of the agentic AI coach was shown to have a stabilizing effect on the decision making process of the administrator, highlighting the importance of its role.