Law enforcement agencies receive a rapidly growing number of reports on online child sexual abuse material. In 2018, the industry, the public and NGOs generated more than 18 million CSAM reports: almost the double of 2017.

These reports are labour intensive to process. The AviaTor project — Augmented Visual Intelligence and Targeted Online Research —, funded by the European Union’s Internal Security Fund – Police, aims at developing automation and intelligence tools to greatly reduce the time spent by LEAs in processing, assessing and prioritizing these reports. These tools combine artificial intelligence techniques, used to investigate the content, with the capability to carry out targeted online research for open source intelligence.

“AviaTor is a tool that is being built for the future.”

Yves Goethals, Judicial Commissioner with Belgian Federal Police.

Our software was developed from a research project into a working prototype in just 12 months. By using expert insight gained from extensive interviews with the Dutch and Belgian police, it was possible to establish how to best process NCMEC referrals.

Read the full article on
AviaTor screenshots

The primary objectives of Project AviaTor

More effective police time and an increase in capacity means more cases handled, more victims rescued, and more offenders caught, creating a safer (European) society that is better able to protect its most vulnerable citizens.

Icon develop

Develop a tool that improves the assessment workflow

Icon reports

Triage relevant reports based on content and characteristics

Icon crawl

Create a tool that can automatically crawl online sources for additional information for investigations in accordance with the national legal requirements

Project AviaTor partners

The members of the consortium for Project AviaTor consist of a mix of technological partners, law enforcement agencies, non-profit organisation and an academic partner.

The National Police of the Netherlands and Police Fédérale Belge are the LEAs that provide information and outline the current scenario of processing, assessing and investigating CSAM reports. Based on this, ZiuZ Forensics and Web-IQ will work in conjunction to develop tools supported by artificial intelligence to improve the workflow. This process will be validated by DFKI, the research partner. The tools developed will first be tested and evaluated by the project partner LEAs and then rolled out to other LEA with the help of the INHOPE network of hotlines and project partner LEAs.

We want to hear from you!

If you are a LEA representative and are interested in learning more about this project and would like to engage with our partners, please contact us by sending an email to

Our technological partners will reach out to you in order to better understand the current landscape of processing CSAM reports in your LEA.

With a higher participation of LEA in describing their current and desired work process, Project AviaTor can build a more robust tool to accommodate the different nuances of national LEA processes and thus help LEA in prioritising reports, focusing on rescuing victims and capturing offenders.

Contact us