Year: 2025
Type: article
Abstract: This paper proposes an AI-driven cybersecurity framework designed to enhance real-time threat detection, reduce response time, and support continuous compliance monitoring. The framework integrates classical machine learning techniques for anomaly detection, reinforcement learning for adaptive remediation, and a compliance engine to align with regulatory standards such as GDPR and NIST CSF. A human oversight interface ensures transparency and accountability in AI-driven decisions. Unlike traditional security tools that operate in isolation, the proposed framework bridges operational security and governance functions, offering a unified approach to cybersecurity management. The framework is described in detail and positioned for implementation in simulated enterprise environments. This work contributes to the field of IT governance and risk management by offering a scalable, regulation-aware model for AI-enhanced security automation.
Authors: Esnaashari, Shadi, Jabal, Mans
Institution: Johns Hopkins University
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7017002061
FWCI:
Cites:
Cited by:
Related to: