Download PDF

Abstract

This capstone project explores the transformation of Security Operations Centers (SOCs) through an AI-driven human–machine collaboration paradigm. Current SOCs face challenges such as alert fatigue, false positives, skill shortages, and inefficiencies caused by tool fragmentation. To address these issues, the project introduces a novel AI architecture based on Multiplex Graph Neural Networks (MGNNs), enabling advanced alert generation, triage, incident response, and intelligent reporting. The framework redefines the role of SOC analysts into AI-driven SOC managers, where human oversight ensures validation, accountability, and continuous improvement of AI systems. By integrating task deconstruction, adaptive learning, and cognitive bias reduction, the proposed model enhances decision accuracy, responsiveness, and scalability. The project emphasizes not only technological innovation but also organizational strategies for adoption, including stakeholder engagement, training, risk management, and ethical considerations. Ultimately, this paradigm advocates for a radical reengineering of SOC operations—democratizing access, reducing costs, and strengthening cybersecurity resilience for enterprises of all sizes.

Author: JACOBS Ivan