University of Moratuwa Undergraduate Research Team Wins Merit Award at SLAAI Best AI Undergraduate Project Awards 2026

The  Merit Award at the SLAAI Best AI Undergraduate Project Award 2026

An undergraduate research team from the Department of Computer Science and Engineering, University of Moratuwa, received a Merit Award at the SLAAI Best AI Undergraduate Project Award 2026, held on 20 November 2026 as part of SLAAI-ICAI 2026. The award recognizes the team's research contribution to addressing privacy challenges in behavioural biometric authentication.

The award-winning project, titled “Identity by Design, Demographics by Accident: Demographic Leakage and Suppression in Behavioral Biometric Embeddings,” investigates an important and often overlooked privacy vulnerability in behavioural biometric systems.

Technologies based on gait, eye movements, keystroke dynamics, and voice are increasingly used for user authentication. Although these systems are designed to recognize individuals, the research demonstrates that their learned identity representations can also encode sensitive demographic information, which can potentially be recovered even without access to the original biometric signals or underlying authentication models.

The study further investigates techniques for suppressing demographic information while preserving authentication performance, highlighting the challenges of developing privacy-preserving biometric systems.

The research was conducted by undergraduate students M. Iyadh Khan, A.P. Patric Nilackshan, M.N.M. Aathif, and L.M. Theesan, under the supervision of Dr. Sandareka Wickramanayake, Department of Computer Science and Engineering, University of Moratuwa, and Dr. Sanka Rasnayaka, National University of Singapore.

The research highlights the importance of evaluating AI systems beyond conventional performance measures. As AI technologies become increasingly integrated into sensitive applications, including biometric authentication, healthcare, and financial services, it is essential to examine their implications for privacy, fairness, security, and accountability.

This work contributes to the research activities of the Trustworthy AI Research Group at the Department of Computer Science and Engineering, University of Moratuwa.

The research paper is available at: Identity by Design, Demographics by Accident: Demographic Leakage and Suppression in Behavioral Biometric Embeddings.

The award was presented by the Sri Lanka Association for Artificial Intelligence (SLAAI), in collaboration with the IEEE Computational Intelligence Society Sri Lanka Chapter.