Disclosure notice: This is a true account of a frontline investigation – written in collaboration with Semantics 21. This may contain accounts that some readers find upsetting.
This case study is based on a real experience shared by a law enforcement/investigation agency professional and written in collaboration with Semantics 21. It is presented in a first-person format to reflect the original voice and lived reality of the investigator, with all identifying information removed or adapted in accordance with UK GDPR and safeguarding standards.
Introduction
A device was submitted for analysis during a routine investigation into a suspect already known to police. The files on the laptop appeared to contain only adult content, primarily selfies and solo videos of the suspect, spread across several folders with timestamp metadata stripped.
An initial manual triage identified nothing obviously illegal. The officer noted the unusual quantity of near-identical videos and stills, but due to the lack of filenames or suspect thumbnails, nothing stood out. Given the suspect’s history, the case was passed to digital forensics for a second review using S21 LASERi-X.
S21 LASERi-X didn’t flag the file because of the main content, it flagged it because of the context around it. That’s what we needed.
Once the media was ingested into S21 LASERi-X, the offline CSAM scan flagged a cluster of seemingly legal adult clips for deeper review. While no direct matches were detected, semantic analysis noted subtle inconsistencies, exposure patterns, staging elements and contextual red flags that hinted something was wrong.
We activated Flickerbook Review to step through the flagged sequence frame by frame. That’s when we saw it: a blurred reflection in a bathroom mirror. Behind the suspect, barely visible through a doorway, was the partial shape of a child.
From Reflection to Revelation
We ran visual enhancement tools, including deblurring, facial upscaling and super resolution, to clarify the reflection. It was enough. Age detection flagged the child as underage, and in a neighbouring frame, a small section of clothing with a partially visible logo gave us our next lead.
Linked media had flagged within the S21 GAD, which confirmed a known victim, tied to an open case in another force area. Our suspect, previously unlinked, was now connected directly to that individual.
A mirror, a blur, a name and a connection no one had made before.
The footage had already been cleared once by manual review but the combination of semantic insight, frame-level analysis, and targeted enhancement brought forward something no human eye caught at speed.
New charges were filed, the safeguarding team for the victim was alerted and a multi-force collaboration was initiated to align the linked investigations.
One reflection, one logo and the chain of proof that turned background information into a breakthrough.
“There was nothing obvious. You’d think it was a typical adult folder. But S21 LASERi-X saw what we missed, a child in the mirror. That moment changed the whole case.”
What if we hadn’t acted?
What if we’d relied on thumbnails? Or trusted that ‘solo content’ meant what it said? What if we’d skipped the mirror entirely?
There was no warning in the filename. No alert from the metadata. Just something quietly wrong in the background.
S21 LASERi-X didn’t just catch a file. It changed what we thought the case was about.
And because of that, someone who had been invisible got seen.
S21 solutions mentioned
S21 LASERi-X
The complete solution for rapid victim identification, CSAM categorisation and media review
Share your experience, inspire others
Have a story of your own? Whether you’d like to be credited or remain anonymous, we’d love to hear from you. Your experience could help rescue a victim, safeguard a child, or take a dangerous offender off the streets. Please get in touch — your words could make a difference.
Request a demo or sales information pack
Please complete the form with valid company or agency information, including a company or agency-issued email address. We will need to confirm your credentials before issuing a free trial licence.