A Concordia-led team of researchers has developed a new artificial intelligence-based method of detecting toxic online content that is faster and more accurate than existing tools. The system is designed to ensure social media platforms can reliably prevent user-generated content they deem harmful from appearing online.