Preventing Online Violence Against Children Using AI

Background

Children around the world face an increasing risk of online violence, with millions of children exposed to harassment and exploitation. This threat has intensified as the COVID-19 pandemic closed schools and moved education and communication online. Save the Children, a global humanitarian organization, has been at the forefront of protecting children for over 100 years. Their new challenge was addressing the growing online violence against children, a new form of harm that requires urgent intervention through advanced technologies like AI.

Objective

The main objective of this project was to develop AI-driven solutions to prevent and mitigate online violence against children. The goal was to use AI to analyze patterns in online behavior, detect potential threats, and build a safer digital environment for children globally. This aligns with Save the Children’s broader mission to protect children from all forms of violence, with a focus on tackling the rising tide of online harassment.

Approach

The team worked in collaboration with the Technology for Development and Child Protection teams at Save the Children. Over the course of two months, we analyzed a variety of data sources including online forums, social media platforms, news articles, and online games. Using Natural Language Processing (NLP) models, the team was able to better understand and categorize the patterns of online violence against children. The models also helped in identifying potential predators and understanding the drivers of online abuse. This comprehensive approach allowed the team to develop a range of technical solutions to predict, detect, and prevent online violence, contributing to Save the Children’s mission to safeguard children’s welfare in the digital space.

Results and Impact

The project resulted in significant insights into the severity, causes, and key actors involved in online violence against children. By analyzing social media, news, and other digital platforms, the team identified the key drivers of online abuse and worked on modeling the behaviors of online predators. The project provided valuable data that can help inform future safety measures and contribute to ongoing efforts to protect children from online harm. The solutions developed during this project are expected to be critical in addressing the rapidly increasing issue of digital violence and exploitation.

Future Implications

The findings from this project have broad implications for future research and policy. By leveraging AI and data analytics, future initiatives could continue to refine and expand methods for detecting online violence, predicting potential threats, and providing interventions in real-time. Furthermore, the outcomes can be integrated into larger-scale global efforts to achieve the UN’s Sustainable Development Goal 16.2, which aims to end all forms of violence against children. These findings could also influence the development of better online safety policies and further research on the ethical use of AI to protect vulnerable populations, particularly children, in the digital age.