Tackling child sexual abuse online with the Safety Tech Challenge Fund

Alex Cross 3 min read

We’re delighted to announce that the Safety Tech Challenge Fund 2021 has awarded Yoti and our collaborators with not one, but two rounds of funding. Working in partnership with Galaxkey & Image Analyzer and DragonflAI, we are looking to tackle child sexual abuse and demonstrate how end-to-end encryption and AI-powered image analysis could be used to create secure solutions that protects young people online.

Between 2019 and 2020, over 10,000 cases of online child sexual abuse were reported to police forces across the UK. In efforts to create change to prevent such crimes, the UK Government created The Safety Tech Challenge Fund. It is a progamme that supports the development of innovative technologies to help provide better protection to victims of child sexual abuse. Led by the DCMS and Home Office, winners have been awarded with £85,000 for their proposals. 

Yoti is delighted to be:

 

1.  Partnering with Galaxkey & Image Analyzer to create an end-to-end secure messaging platform

End-to-end email encryption platform, Galaxkey has partnered with Yoti and content analysis platform Image Analyzer to innovate a messaging platform that detects explicit content before it is sent, rather than after the fact.

The tech would work by creating an encrypted messaging platform that allows the user to verify their age using Yoti’s facial age estimation or the Yoti digital ID app when they create an account. Any messages sent via the platform will go through an explicit content detection process using Image Analyzer’s AI to prevent explicit material from being sent to a recipient. The solution is also expected to evolve beyond images in the future, scanning voice, text and URLs. We hope this will provide even more protection against explicit content. Upon detection, the content would be sent to the reporting server on Galaxkey’s NCSC accredited encrypted workspace platform for moderation.

By creating a working platform which can detect, action and report explicit content prior to delivery of the content to a recipient, we hope to prevent child sexual abuse material from being sent via messaging platforms.

 

2. Working with DragonflAI to detect and flag child sexual content on a device

We’ve also joined forces with video and image moderation company DragonflAI to create a solution that instantly detects a person’s age in explicit content on a device and completely offline.

Specialising in the detection of nudity, DragonflAI’s algorithm has a success rate of 97.9%. Combined with Yoti’s age estimation technology, the collaboration will be able to detect the ages of people involved in explicit content. This will then allow successful moderation of illegal content without a third-party being involved. 

The goal of this collaboration is to create a tool that can detect unseen and new indecent images with a high degree of accuracy without the content leaving a device.

We are proud to be putting our solutions forward to encourage innovation, helping change the digital space to better protect children online. We thank the Safety Tech Challenge Fund for welcoming the use of tech to tackle the rise in online-linked sexual crimes, and look forward to working with our partners to create tools that make the internet a safer place for children.