ATHENA - Addressing Hate Speech Through Empathic Virtual Humans and AI

Hate speech poses serious threats to both its immediate targets and democratic societies across Europe, especially with recent AI-driven technologies increasing its reach and impact. Despite the technical progress in the detection and analysis of hateful content, understanding the psychological and social mechanisms underlying how individuals perceive hate speech, those initiating it and their targets, remains limited. ATHENA (Addressing Hate Speech Through Empathic Virtual Humans and AI) brings together a European network of researchers seeking to uncover hate speech as a multimodal phenomenon and develop interventions utilising recent advancements in Virtual Humans (VHs), i.e., hyper-realistic, embodied, digital, intelligent agents, and Artificial Intelligence.
The consortium comprises Breda University of Applied Sciences (lead), with expertise in virtual human and AI technologies, and the University of Bremen and the University of Warsaw, with expertise in psychophysiology, social psychology and particularly hate speech. The partners will lead collaborative efforts to build the foundation for a Horizon Europe application targeting call HORIZON-CL2-2027-01-DEMOCRACY-04: Addressing the impact of artificial intelligence, cyberviolence, and deepfakes on equality, democracy and inclusive societies. Some of the activities preceding the development of the final application include assessing the current consortium's capacity, identifying and recruiting new partners, and conducting a proof-of-concept study to generate strategic preliminary data for the final proposal.
The interdisciplinary network bridges Dutch regional digital realities ambitions with the innovation capacity of EU researchers, strengthening Dutch participation in addressing societal challenges in European consortia. The project aligns with national and European priorities in AI ethics and digital societies by focusing on urgent societal challenges: countering hate speech, protecting vulnerable population groups, and safeguarding democratic values through the responsible deployment of AI.