Forensic Biometrics: the use of biometric data, databases and technology in forensic applications
The second lecture will focus on the use of biometric data, databases and technology in forensic practice. Practical examples will be proposed to explain the forensic biometric applications: intelligence, investigation and evaluation in court.
Forensic biometric practice aims at answering questions about the source of biometric items and about the reconstruction of activities (e.g. type, location, time and sequence) involving biometric items. Concretely the process consists in examining (i.e. recovering, analysing and interpreting) biometric items, and in answering the questions by reporting the observations made and the opinions formulated. Before to be implemented in casework, forensic biometric methods need to demonstrate compliance to quality assurance. They need to be extensively validated with real and realistic forensic biometric data to determine their limits and scope of validity when used by forensic examiners.
The use of forensic biometrics also extends to the examination of biometric items in less traditional conditions, such as conflict zones. In such cases forensic biometrics is requested to answer other relevant questions, beyond the question of the source, including the assessement of the authenticity and integrity of the recovered items and the processing of large amount of data to characterise people or their activities. This new type of requests opens challenges for R&D in terms of relevant datasets, biometric technology, forensic inference as well as legal and ethical aspects.
The rapid development and widespread use artificial intelligence in the society also affects forensic biometrics. The lecture will focus on the generative (GenAI) and discriminative (DisAI) applications derived from artificial intelligence models that play a role in forensic biometrics: On one side the use of GenAI to produce artificial forensic biometric items (e,g. morphing, deepfakes); on the other side, the use of GenAI and DisAI for the forensic recovery, analysis, interpretation and reporting. It will also explores the accountability requirements for forensic biometric processes embedding GenAI and DisAI, from a scientific, legal, and societal perspective.