Authors : Sreenivasa Rao Basavala; Prudhvi Raju Mudunuri

Volume/Issue : Volume 11 - 2026, Issue 5 - May

Google Scholar : https://tinyurl.com/9ahau8ep

Scribd : https://tinyurl.com/py48jaru

DOI : https://doi.org/10.38124/ijisrt/26May925

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Artificial Intelligence (AI) is changing the way organizations work with new technologies that help to enhance the security of their information assets, while also creating new attack vectors. While AI has the potential to dramatically improve an organization’s ability to detect threats, automate repetitive administrative tasks and provide more real-time responsive systems, there are associated risks of exposure, including vulnerabilities in the new systems and software, as well as new types of attack vectors. Examples of new types of AI-based attacks that have been recently discovered and are reported in the research include but are not limited to: - Adversarial attacks - Data poisoning - Prompt injections - Model evasion - Model theft - AI-driven social engineering attacks such as deepfakes and other automated phishing campaigns. These attacks can lead to many types of incidents, including data theft of sensitive data, denial of service, reputational damage due to loss of customer trust and more. At the same time, new attack surfaces have been created, for example, in the form of training data for the new systems, the structure and design of the systems, and dependencies in third party applications and services. This paper aims to provide a deep dive into the many types of cyberattacks that exist in the realm of AI and to provide an in-depth analysis of their methods, techniques and the overall impact of these new types of attacks on the wider Cybersecurity landscape. This paper also aims to give an in-depth look at countermeasures and defenses that can be put in place to help combat these threats, including but not limited to secure coding practices for the development of new systems and AI models, the use of adversarial testing, access controls and real-time threat detection and alerting. Organizations need to be aware of the potential threats of these new technologies and the need to secure their systems using several security controls to mitigate the threats and to be prepared.

Keywords : AI Security, OWASP LLM Top 10, AI Supply Chain Attacks, Prompt Injection Attacks, Model Theft, Poisoning Model, AI Cyberattacks, Mitigation Techniques.

References :

1. Bertino, E., Kantarcioglu, M., Akcora, C. G., Samtani, S., Mittal, S., & Gupta, M. (2021, April). AI for Security and Security for AI. In Proceedings of the eleventh ACM conference on data and application security and privacy (pp. 333-334).
2. Dash, Atish Kumar. "Securing the LLM Supply Chain: Analyzing Threats and Mitigation Strategies." In 2026 IEEE 5th International Conference on AI in Cybersecurity (ICAIC), pp. 1-7. IEEE, 2026.
3. Fakhouri, Hussam N., Basim Alhadidi, Khalil Omar, Sharif Naser Makhadmeh, Faten Hamad, and Niveen Z. Halalsheh. "Ai-driven solutions for social engineering attacks: Detection, prevention, and response." In 2024 2nd international conference on cyber resilience (ICCR), pp. 1-8. IEEE, 2024.
4. Husak, O., Moroz, R., & Denysenko, N. (2025). AI security.
5. Jia, Y., Liu, Y., Shao, Z., Jia, J., & Gong, N. (2025). Promptlocate: Localizing prompt injection attacks. arXiv preprint arXiv:2510.12252.
6. John, S., Del, R. R. F., Evgeniy, K., Helen, O., Idan, H., Kayla, U., ... & Vasilios, M. (2025). Owasp top 10 for llm apps & gen ai agentic security initiative (Doctoral dissertation, OWASP).
7. Kezron, N. "Securing the AI supply chain: Mitigating vulnerabilities in AI model development and deployment." World Journal of Advanced Research and Reviews 22, no. 2 (2024): 2336-2346.
8. Kure, Halima I., Pradipta Sarkar, Ahmed B. Ndanusa, and Augustine O. Nwajana. "Detecting and preventing data poisoning attacks on AI models." In 2025 Photonics & Electromagnetics Research Symposium-Spring (PIERS-Spring), pp. 01-12. IEEE, 2025.
9. Liu, J., Truhn, D., Zhao, Y., Gupta, M., Abera, Y., Ajayi, J., ... & Arif, H. (2025). Overreliance on AI Systems and Skill Degradation Risks Among Operators in Critical Infrastructure Cybersecurity Environments.
10. Morozumi, Arisa, and Hisashi Hayashi. "LLM-based risk scenario generation and mitigation for AI systems: A case study approach." In International Joint Conference on Computational Intelligence, pp. 269-293. Cham: Springer Nature Switzerland, 2025.
11. Parisa, S. K., & Banerjee, S. (2024). Ai-enabled cloud security solutions: A comparative review of traditional vs. next-generation approaches. International Journal of Statistical Computation and Simulation, 16(1).
12. Ragab, N., Ahmed, A., & AlHashmi, S. (2015, June). Software engineering for security as a non-functional requirement. In Intelligent Data Analysis and Applications: Proceedings of the Second Euro-China Conference on Intelligent Data Analysis and Applications, ECC 2015 (pp. 347-357). Cham: Springer International Publishing.
13. Ramirez, M. A., Kim, S. K., Hamadi, H. A., Damiani, E., Byon, Y. J., Kim, T. Y., ... & Yeun, C. Y. (2022). Poisoning attacks and defenses on artificial intelligence: A survey. arXiv preprint arXiv:2202.10276.
14. Reddy, Pavan. "Weaponizing Words: Direct & Indirect Prompt Injection Attacks on LLM." In Proceedings of the 26th ACM Annual Conference on Cybersecurity & Information Technology Education, pp. 292-293. 2025.
15. SAMUEL, A. (2025). Predictive AI for Supply Chain Management: Addressing Vulnerabilities to Cyber-Physical Attacks. Well Testing Journal, 34(S2), 185-202.
16. Tang, Ruixiang, Hongye Jin, Mengnan Du, Curtis Wigington, Rajiv Jain, and Xia Hu. "Exposing model theft: A robust and transferable watermark for thwarting model extraction attacks." In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, pp. 4315-4319. 2023.
17. Tao, G., Cheng, S., Zhang, Z., Zhu, J., Shen, G., Han, W., ... & Zhang, X. (2025, June). A Systematic Threat Modeling of LLM Applications. In Proceedings of the 33rd ACM International Conference on the Foundations of Software Engineering (pp. 1607-1614).
18. Vulchi, Jaswanth Reddy, and Eric Ackerman. "Exploring owasp top 10 security risks in llms with practical testing and prevention." (2024).
19. Wymberry, C., & Jahankhani, H. (2024). An approach to measure the effectiveness of the mitre atlas framework in safeguarding machine learning systems against data poisoning attack. In Cybersecurity and artificial intelligence: Transformational strategies and disruptive innovation (pp. 81-116). Cham: Springer Nature Switzerland.