Adaptive Cybersecurity Threat Intelligence Using Explainable Artificial Intelligence for Resilient Protection of U.S. Critical Information Systems

Abstract

This study examined the relationship among adaptive cybersecurity threat intelligence, explainable artificial intelligence, analyst trust, detection accuracy, response efficiency, and cyber resilience within U.S. critical information systems. The increasing sophistication of cyber threats targeting healthcare, finance, telecommunications, transportation, energy, and governmental infrastructures created significant demand for intelligent cybersecurity systems capable of improving operational resilience, transparency, and response coordination. The study used a quantitative, cross-sectional, correlational research design grounded in adaptive cybersecurity theory, cyber resilience theory, and explainable artificial intelligence theory. Data were collected from 128 cybersecurity professionals working across critical infrastructure sectors in the United States using a structured Likert-scale survey instrument. Descriptive statistics, Pearson correlation analysis, and multiple linear regression analysis were conducted using SPSS, R, and Python statistical software to examine the relationships among the study variables. The findings revealed strong positive relationships among adaptive cybersecurity threat intelligence, explainable AI transparency, analyst trust, detection accuracy, response efficiency, and cyber resilience. Adaptive cybersecurity threat intelligence demonstrated a high mean score of 4.18, while cyber resilience reported the highest overall mean score of 4.22, indicating strong organizational perceptions regarding AI-supported cybersecurity effectiveness. Pearson correlation analysis revealed a strong positive relationship between adaptive threat intelligence and cyber resilience (r = 0.79, p < 0.01), while response efficiency demonstrated the strongest correlation with cyber resilience (r = 0.84, p < 0.01). Multiple linear regression analysis indicated that the independent variables collectively explained 76% of the variance in cyber resilience outcomes (R² = 0.76, p < 0.001). Adaptive threat intelligence emerged as the strongest predictor of cyber resilience (β = 0.432, p < 0.001), followed by response efficiency (β = 0.401, p < 0.001) and explainable AI transparency (β = 0.314, p < 0.001). The findings further demonstrated that organizations with greater levels of explainable AI integration and adaptive cybersecurity implementation experienced stronger operational continuity, improved analyst trust, enhanced threat visibility, and reduced operational disruption during cybersecurity incidents affecting critical information systems.