In the fight against child sexual abuse, law enforcement officers face daunting challenges, not least of which is the overwhelming task of sifting through digital evidence. For any one suspect, investigators might seize phones, laptops, and hard drives containing thousands of files. The forensic processing that follows can be time consuming and even traumatic for officers reviewing evidence, who might find child sexual abuse material (CSAM) mixed among other photos.
That’s why Thorn developed its CSAM Classifier, a machine learning-based tool that can find new and previously unknown CSAM — material that exists but hasn’t yet been classified as CSAM. It’s a critical tool for law enforcement on the front lines of child sexual abuse crimes, allowing them to more quickly find new CSAM, while controlling their exposure to such appalling abuse.
Getting this technology into the hands of as many law enforcement agencies as possible is critical to help aid their investigations into child sexual abuse cases active abuse That’s why we’re excited to announce that through our partnership with Griffeye, the Sweden-based world leader in digital media forensics for child sexual abuse investigations, Thorn’s CSAM Classifier is now available directly in Griffeye Analyze, a platform used as a home base by law enforcement worldwide.
The integration of Thorn’s CSAM classifier into Griffeye’s Analyze DI Pro platform marks a pivotal moment in the ongoing battle against child sexual abuse. This collaboration brings together Thorn’s advanced technology with Griffeye’s extensive reach within the law enforcement community, creating a formidable toolset designed to cut through the overwhelming volume of data to uncover victims of abuse more swiftly and effectively than ever before.
Griffeye Analyze helps agencies manage, categorize, and match large volumes of images and videos to detect illegal activity, especially child sexual abuse. With the integration of CSAM Classifier, Griffeye Analyze becomes an even more comprehensive tool that elevates unknown CSAM images and video for triage, review, and escalation.
Thorn’s CSAM Classifier uses machine learning trained on historical data to determine which photos and videos are most likely to be CSAM. When it identifies new or previously unknown CSAM, it flags the file and a moderator reviews it to confirm. The tool learns from this feedback and over time improves its ability to spot new CSAM, automating what used to be an impossibly difficult manual task. It also eliminates duplicate efforts by classifying CSAM for others to identify and track. That means investigators can increase their capacity to process and solve cases.
At the same time, the CSAM Classifier helps protect the well-being of these investigators, who are on the front lines of some of the world’s most horrific crimes. The tool acts as a filter that controls their exposure to CSAM so they can pursue cases while limiting the mental health impacts of encountering such horrific abuse.
In our partnership with Griffeye, we’re expanding our impact by providing law enforcement with better tools that create a stronger, more unified, and more resilient front against child sexual abuse.