Kaspersky has introduced a new professional training program titled “Large Language Models Security,” designed to prepare cybersecurity specialists for the emerging challenges surrounding AI and large language models (LLMs). Developed by the Kaspersky AI Technology Research Center, the course focuses on identifying, assessing, and mitigating vulnerabilities unique to LLM-powered systems—a growing concern as organizations increasingly integrate AI into their operations.
According to Kaspersky’s own research, by 2024 more than half of global companies had already implemented AI and Internet of Things (IoT) technologies in their infrastructure. This rapid adoption has expanded the attack surface for cybercriminals, making AI security a critical skill rather than a niche specialization. The new training program addresses this demand by equipping professionals with both theoretical knowledge and hands-on experience in defending AI systems from targeted threats.
The course, led by Vladislav Tushkanov, Research Development Group Manager at Kaspersky, blends video lectures, lab-based exercises, and real-world case studies to help participants understand how LLMs can be exploited and how to safeguard them. Through interactive modules, trainees will learn to:
- Examine common exploitation methods such as prompt injections, jailbreaks, and token smuggling to understand their mechanics.
- Develop defense strategies at multiple levels—covering the model, prompt, system, and service layers.
- Use structured frameworks for assessing risk and improving the resilience of AI-driven platforms.
The program is designed for a broad range of participants, including cybersecurity newcomers, AI engineers, and IT professionals integrating LLMs into their infrastructure. By the end of the course, participants are expected to have a clear understanding of how to evaluate AI model vulnerabilities, deploy layered defense systems, and ensure the long-term integrity of LLM-based applications.
Tushkanov describes the motivation behind the course as a response to the evolving digital landscape: “The rise of large language models has reshaped how organizations engage with AI, but it also brings complex security puzzles that demand urgent attention. Our goal is to help professionals master the techniques needed to protect LLM-driven systems and stay ahead of emerging threats.”
This addition strengthens Kaspersky’s growing Cybersecurity Training portfolio, which includes specialized programs aimed at developing advanced, practical skills for industry professionals. As part of the company’s long-standing investment in secure AI research, the new LLM-focused course reflects nearly two decades of Kaspersky’s experience in applying artificial intelligence to threat detection and prevention.
Enrollment for the Large Language Models Security training is now open, with details and registration available through Kaspersky’s official website.
