AI4Dignity is a proof of concept project funded by the European Research Council (2020-2022) that aims to address challenges facing AI-assisted extreme speech moderation. Such challenges include ensuring the quality, scope and inclusivity of training data sets and developing AI that accounts for context and cultural nuances. These points are compounded by the overarching challenge in defining what constitutes extreme speech and hate speech. To address these, AI4Dignity seeks to develop a tool for content moderation that moves beyond keyword-based detection systems. To do this, AI4Dignity will pioneer a collaborative and community-based classification approach. In this way, AI4Dignity seeks to take a significant step towards setting procedural benchmarks to operationalize “the human in the loop” principle and bring inclusive training datasets for AI systems tackling urgent issues of digital extreme speech.

For more updates and information, visit the project website.