Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{174337,
        author = {Chandana D and Kunjal Singh and Jayam Bhavagna and Guru Prasath and Simran Gupta and Mrs Shilpa Mary and Dr Sachin.K.Parappagoudar},
        title = {A study on Comprehensive AGI Threat modelling: Cybersecurity approach to safeguarding AGI systems},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {10},
        pages = {3529-3540},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=174337},
        abstract = {Artificial Generative Intelligence is taking over every major aspect of daily lives and business operations. The risk of AGI’s capacity to think, adapt, and make judgments has made AGI to evolve to be AI Agent – which is more capable of acting on its own across a variety of disciplines, with this flexibility brings with it previously unheard-of cybersecurity risks, from malicious intent and systemic failures to adversarial exploitation and self-modifying code vulnerabilities. This paper attempts to study the AGI threat modelling using cybersecurity approach to ensure safe AGI system usage. By leveraging the cybersecurity community’s efforts and science of threat detection and mitigation in this paper. This research looks at how different dangers associated with AGI systems are identified and categorized. Adversarial assaults, self-modifying code flaws, malicious actor abuse, and systemic breakdowns resulting from unforeseen actions are some of these hazards. By applying cybersecurity principles and examining existing threat modelling frameworks, this study offers a multi-layered protective solution to enhance AGI security. To minimize risks, the proposed approach prioritizes proactive measures including robust encryption, adversarial training, continuous monitoring, and ethical AI governance. Our research aims to foster trust and reliability in AGI technology while contributing to the development of morally sound AGI systems that can be securely integrated into society.},
        keywords = {Artificial Generative Intelligence, AI-Agents, Cyber security, AI-Risks, AGI Threat modelling, AI Risk Management, Ethical AI Security, Machine Learning Security.},
        month = {March},
        }