AI for Good Hackathon: AI Against Modern Slavery

October 14-28, 2020

In 2018, the Global Slavery Index found that there were 40.3 million people in modern slavery, of whom 25 million were in forced labor producing computers, clothing, agricultural products, raw materials, etc. and 15 million were in forced marriage. In order to facilitate measures to end modern slavery, the United Nations encourage governments and companies to take immediate measures. 

As a result of newly implemented national legislation in the UK and Australia, large commercial organizations are now legally required to publish statements highlighting their efforts to investigate and take action against modern slavery in their supply chains. However, assessment of these companies’ statements is manual and time consuming. Cutting-edge machine learning methods can dramatically speed up analysis of companies’ statements to help governments and the industries at large to hold them accountable.

There are now tens of thousands of modern slavery statements published by global companies from across industries. As part of Project AIMS, The Future Society has begun curating a machine readable dataset of these text statements to enable machine learning for analysis. 

Among many initiatives that companies can undertake to eradicate modern slavery, providing training and capacity building about slavery and human trafficking to employees, leadership and suppliers is highly effective. As part of the TWIML AI virtual festival, The Future Society is organizing a hackathon challenging machine learning practitioners and AI enthusiasts to develop accurate text classification algorithms to distinguish which companies’ modern slavery statements discuss the provision of training and capacity building for their company and suppliers.  

By sharing your analysis and contributing to this repository you help the global community to hold multinational corporations accountable for how they treat their workforce and suppliers.

Watch the introduction of the session here.