Research

Dr. Klisura’s research lies at the intersection of artificial intelligence and cybersecurity, with a focus on trustworthy and responsible AI systems. His work examines how large language models and other AI systems can be made more secure, reliable, fair, and suitable for deployment in sensitive and regulated environments.

A central theme of his research is understanding how AI systems behave when they mediate access to sensitive information. His work has investigated confidentiality risks in text-to-SQL systems, access-control reasoning in large language models, dialect-related disparities in AI safety and privacy applications, and methods for improving the security and fairness of AI systems. His broader cybersecurity research also includes enterprise network security, IoT malware analysis, and data-driven security analytics.

His current research interests include:

• Trustworthy and Responsible Artificial Intelligence
• Large Language Models and Natural Language Processing
• AI Security, Privacy, and Safety
• Cybersecurity
• Data Analytics and Machine Learning

His research has been published in venues including the Annual Meeting of the Association for Computational Linguistics (ACL), NAACL, EACL, the IEEE Symposium on Security and Privacy, and IEEE Transactions on Dependable and Secure Computing.

For a complete list of publications and citation information, please visit Dr. Klisura’s Google Scholar profile.