Automating CSAM Investigation: Research to Practice

Notes:

1. The software developed by Bryce Westlake, Russell Brewer and colleagues aims to assist law enforcement agencies in identifying perpetrators of child sexual abuse material (CSAM) offences by using multiple biometric markers.
2. The unique function of the software is its capability to synthesize, triage, and systematically review all evidence holdings that law enforcement might seize, thereby automating the process and speeding up investigations.
3. The software can ingest a variety of media files, extract multiple biometric features from those files, and identify subjects contained within, matching them across data sets, reducing investigator exposure to CSAM and speeding up the investigative process.
4. One of the key aspects addressed by the software is the ability to identify new content, overcoming the limitations of current reactive methods used in combating CSAM (i.e. matching file hashes).
5. The software is leverages social networking analysis and other techniques to assist investigators to identify links between offenders, victims, places and objects, potentially opening new avenues or focusing detective work.
6. The project aims to connect law enforcement agencies nationally and globally, enabling them to share information and collaborate on cases, despite the challenges posed by different laws and data transmission regulations.
7. The software's ability to systematically catalog and review all available data provides a more comprehensive and unbiased investigative process as choices about which media is reviewed, reducing the impact of investigator intuition and fragmented intelligence.
8. The software has received funding for research and development, leading to the creation of a functional prototype for law enforcement in Australia and they are now looking at having it implemented by police more widely.
9. The development of AI technology has a positive impact on the technology, as new matching tools are "plug and play".
10. One of the challenges faced in implementing the software is convincing law enforcement to adopt new technologies and methods, as they may be hesitant to change from established routines, despite the potential benefits of the new system.

About our guests:

Russell Brewer

https://researchers.adelaide.edu.au/profile/russell.brewer#professional-activities

Bryce Westlake

https://www.brycewestlake.com

https://www.sjsu.edu/justicestudies/about-us/directory/westlake-bryce.php

Papers or resources mentioned in this episode:

Brewer R et al. 2023. Advancing child sexual abuse investigations using biometrics and social network analysis. Trends & issues in crime and criminal justice no. 668. Canberra: Australian Institute of Criminology. https://doi.org/10.52922/ti78948

Westlake B et al. 2022. Developing automated methods to detect and match face and voice biometrics in child sexual abuse videos. Trends & issues in crime and criminal justice no. 648. Canberra: Australian Institute of Criminology. https://doi.org/10.52922/ti78566

DEVELOPING AUTOMATED SOFTWARE TOOLS: To Detect Child Sexual Abuse Material Online.

https://adelaidecybercrime.org/software

Other:

This interview