Emerging LLM Threats: A Comprehensive Analysis of Attacks and Mitigation

Sreenivasa Rao Basavala () and Prudhvi Raju Mudunuri ()

International Journal of Innovative Science and Research Technology (IJISRT), 2026, vol. 11, issue 05, 2136-2145

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. (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.ijisrt.com/emerging-llm-threats-a-comp ... tacks-and-mitigation (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:cvr:ijisrt:2026:05:ijisrt26may925

DOI: 10.38124/ijisrt/26May925

Access Statistics for this article

More articles in International Journal of Innovative Science and Research Technology (IJISRT) from IJISRT Publication
Bibliographic data for series maintained by Rahul Goyel ().