Leveraging Generative AI and Behavioral Biometrics to Strengthen Zero Trust Cybersecurity Architectures in Healthcare Systems
Akinde Michael Ogunmolu
*
Texas A&M University, 700 University Blvd, Kingsville, TX 78363, United States.
*Author to whom correspondence should be addressed.
Abstract
This study investigates the integration of Generative Artificial Intelligence (AI) and Behavioral Biometrics within Zero Trust cybersecurity frameworks tailored for healthcare systems. Drawing on public datasets—including MITRE ATT&CK for Enterprise, UNSW Behavioral Biometrics, and the HHS HC3 Healthcare Cybersecurity Dashboard—the study employs a quantitative design featuring Two-Sample T-Tests, ROC-AUC analysis, and Exploratory Factor Analysis (EFA). Findings indicate a marked improvement in threat detection rates, rising from 65.21% to 82.06% following Generative AI deployment. Additionally, behavioral biometric authentication models achieved an Area Under the Curve (AUC) score of 1.00, signifying perfect classification performance. While these results highlight the technical promise of both technologies, such outcomes are derived under controlled experimental conditions using curated public datasets. The AUC score of 1.00, while mathematically accurate within the testing dataset, may not fully translate to real-world clinical environments due to potential overfitting, limited behavioral variability, and the absence of adversarial behavior simulations. These limitations underscore the need for caution when generalizing findings to dynamic healthcare infrastructures, where biometric stability and adaptive threat detection are more variable. The study further identifies financial, infrastructural, and policy-level challenges as critical barriers to adoption. Strategic recommendations include investing in interoperable cybersecurity infrastructure, updating HIPAA frameworks to address AI and biometrics governance, expanding interdisciplinary education programs, and introducing policy incentives for Zero Trust transformation. Recommendations include investing in interoperable infrastructure, updating HIPAA frameworks, expanding interdisciplinary cybersecurity education, and introducing governmental incentives to support Zero Trust transformations enhanced by AI and biometrics.
Keywords: Generative AI, behavioral biometrics, zero trust architecture, healthcare cybersecurity, exploratory factor analysis