top
The logo of Projekt Arachnid shows a stylised spider.

The hash database with digital fingerprints of images is used for an automated identification of known child sexual abuse material.

Project Arachnid

The logo of Projekt Arachnid shows a stylised spider.

The hash database with digital fingerprints of images is used for an automated identification of known child sexual abuse material.

Automated combatting of child sexual abuse material

Since May 2020, jugendschutz.net takes part in Project Arachnid for the identification of known child sexual abuse material. The core of the system developed by the Canadian Center for Child Protection is a hash database with digital fingerprints of images used for an automated identification of known child sexual abuse material. Using the Arachnid database, an automated tool checks images displayed on reported URLs and matches them with the database hash values. This shall ensure that known CSAM content can no longer be disseminated, protecting not only the victims but also analysts who don't have to repeatedly examine identical content.

As soon as Arachnid identifies known illegal content, the system automatically informs law enforcement, partner hotlines or providers hosting the content.

Classification by participating analysts

Currently, Arachnid deals with explicit child sexual abuse material, banned throughout the world. The analysts of the organizations participating in the project manually classify the images hashed. The hash value based on a classification is only complete if three analysts agree on the assessment.

Arachnid only matches data to content that was reported to the participating hotlines as child sexual abuse material. Arachnid also crawls links on sites reported that contained child sexual abuse material. As soon as the system finds no more illegal content, it stops crawling. This make it clear that Arachnid is not a monitoring tool censoring the web; the activities are solely based on reasonable suspicion.

It is open to ISPs, platform operators and hosting providers to compare the content in their services to the Arachnid database and thus prevent the upload of illegal content in the area of child sexual abuse material beforehand. Here, it is not necessary to look at the data since Arachnid only compares hash values.

The automated processing of known content frees up resources to deal with new, unknown material and makes the hotline work more effective. Analysts can concentrate on those reports that are not processed automatically and thus focus on the most current cases.