IRMA-International.org: Creator of Knowledge
Information Resources Management Association
Advancing the Concepts & Practices of Information Resources Management in Modern Organizations

Multi-Agent Generative AI Systems for Cyber Risk Evaluation and Control

Multi-Agent Generative AI Systems for Cyber Risk Evaluation and Control
View Sample PDF
Author(s): Syed Mohd Faisal (Malla Reddy Technical Campus, Malla Reddy Vishwavidyapeeth (Deemed to be University), Hyderabad, India), Wasim Khan (Symbiosis Institute of Technology, Symbiosis International (Deemed to be) University, Pune, India.), Mohammad Ishrat (VIT Bhopal University, India)and Anwar Ahamed Shaikh (Sanjivani University, Kopergaon, India)
Copyright: 2027
Pages: 34
Source title: Generative AI for Cyber Risk Management
Source Author(s)/Editor(s): Yassine Maleh (Sultan Moulay Slimane University, Morocco), Lahby Mohamed (Hassan II University, Casablanca, Morocco)and Ahmed A. Abd El-Latif (Prince Sultan University, Saudi Arabia)
DOI: 10.4018/979-8-3693-8397-1.ch003

Purchase

View Multi-Agent Generative AI Systems for Cyber Risk Evaluation and Control on the publisher's website for pricing and purchasing information.

Abstract

We present a systems-theoretic framework coupling Generative AI with multi-agent systems to make cyber-risk management a continuous, anticipatory control loop. The architecture: (i) risk-evaluation agents fusing telemetry with a knowledge graph; (ii) generative threat agents (GANs/diffusion/LLMs) synthesizing counterfactual attacks; and (iii) control agents trained with CTDE to allocate defenses under safety constraints. We implement a closed-loop simulation and evaluate ransomware and IoT-botnet cases using AUROC/AUPRC/F1 and time-to-containment. Ablations isolate contributions of generative foresight, coordination, and graph context. Results: higher detection fidelity and faster containment than strong non-generative/non-agentic baselines. We discuss governance, limitations (sim-to-real, MARL stability), and directions in continual learning, risk-sensitive control, and human–AI teaming.

Related Content

Frederic Andres. © 2027. 14 pages.
Kalsoom Safdar, Khairul Najmy Abdul Rani, Mohd Aminudin Jamlos, Siti Julia Rosli, Muhammad Usman Younus, Zanab Safdar. © 2027. 27 pages.
Bani Adam, Binastya Anggara Sekti, Muhammad Adi Zacky Zahran. © 2027. 24 pages.
Swetha Margaret T. A., Renuka Devi D.. © 2027. 31 pages.
Maurice Saluschke, Michael Schulz. © 2027. 30 pages.
Mirjam Sepesy Maučec, Gregor Donaj. © 2027. 16 pages.
Jorge A. Ruiz-Vanoye, Ocotlan Diaz-Parra, Ricardo A. Barrera-Cámara, Alejandro Fuentes-Penna, Francisco R. Trejo-Macotela, Jaime Aguilar-Ortiz, Eric Simancas-Acevedo. © 2027. 21 pages.
Body Bottom