CyberGaar conducted an AI Red-Team Workshop at the University of Karachi’s UBIT on November 2, 2026, bringing together around 90 students to explore emerging cybersecurity challenges linked to artificial intelligence. The session introduced participants to practical AI security concepts and examined how AI technologies are influencing vulnerability research, penetration testing and security assessments. As AI-powered applications become increasingly common across industries, understanding their potential weaknesses and security requirements is becoming an important part of cybersecurity education. The workshop aimed to familiarise students with emerging attack techniques and the considerations involved in assessing and protecting AI-based systems. By bringing these discussions into an academic setting, CyberGaar provided participants with an opportunity to learn about developments in AI security and the technical knowledge required to address associated risks. The event also reflected the organization’s focus on preparing aspiring cybersecurity professionals to work with modern technologies while maintaining a strong foundation in established security practices. The participation of approximately 90 students highlighted growing interest in the changing cybersecurity landscape and the role of AI security research in protecting digital systems.
A major component of the workshop focused on the OWASP Top 10 for Large Language Model (LLM) Applications, a framework that outlines key security risks affecting applications built around large language models. Students were introduced to vulnerabilities that can emerge in LLM-powered systems, along with practical concepts for identifying and understanding these weaknesses. The discussion explored emerging attack techniques and the importance of assessing AI applications from a security perspective. As organizations increasingly integrate large language models into their products and digital operations, security researchers need to understand the risks associated with these technologies and how weaknesses may be identified or exploited. The workshop highlighted the importance of recognizing potential vulnerabilities, evaluating their impact and understanding the security considerations involved in developing and deploying AI-powered applications. Introducing the OWASP framework also gave participants a recognized reference for exploring risks specific to LLM applications. The session demonstrated why AI security is becoming an increasingly relevant area of professional development and why future cybersecurity practitioners need to build the skills required to assess emerging technologies responsibly.
The workshop further examined how AI-powered tools are changing vulnerability research and penetration testing. These technologies can support security professionals in examining applications, identifying potential weaknesses and carrying out security assessments. However, effective security research still requires technical expertise, careful validation and a responsible approach to testing. CyberGaar emphasized the importance of developing skilled researchers who can identify, verify and address emerging threats while understanding the limitations of automated tools. Participants were encouraged to combine traditional cybersecurity fundamentals with the capabilities offered by modern AI technologies. Researchers must be able to distinguish genuine vulnerabilities from inaccurate findings, assess the significance of identified weaknesses and understand their potential impact before drawing conclusions. This balance is particularly important as AI becomes more widely used in security research and application development. By introducing students to both established security practices and AI-assisted techniques, the workshop aimed to help them understand the evolving requirements of the profession. The knowledge gained through such sessions can support aspiring researchers in assessing conventional digital infrastructure as well as newer applications built on artificial intelligence.
CyberGaar thanked UBIT, University of Karachi, for hosting the workshop and acknowledged the participating students for their enthusiasm and interest in learning about AI security. The organization reiterated its commitment to student education, cybersecurity awareness, innovation and the development of startups, describing these areas as important investments in the future of the digital ecosystem. The event connected academic learning with practical discussions on AI security, vulnerability research and penetration testing, giving students greater exposure to an evolving area of cybersecurity. As artificial intelligence continues to influence digital products and services, organizations will need professionals who can understand the security implications of these systems and help identify and address potential weaknesses. Providing students with opportunities to explore emerging technologies can contribute to the development of relevant technical skills and encourage further research and innovation. Through initiatives such as the AI Red-Team Workshop, CyberGaar aims to support the next generation of cybersecurity professionals in combining established security expertise with modern AI assessment capabilities. The workshop reinforced the value of preparing students to recognize and respond responsibly to the security challenges associated with the growing adoption of artificial intelligence.
Follow the SPIN IDG WhatsApp Channel for updates across the Smart Pakistan Insights Network covering all of Pakistan’s technology ecosystem.
