Project
Detecting and Countering Hate Speech Targeting Women in Armenia
Duration
October 2025 – May 2026 (8 months)
Project Goal
To design, develop and pilot an AI-based platform that detects and helps counteract hate speech targeting women in the Armenian digital environment.
Context
Online hate speech against women is a growing challenge in Armenia’s digital landscape. Current moderation and detection mechanisms are often inadequate in Armenian, Latin, Cyrillic, and transliterated texts. Civil society, media outlets, and public figures lack tools tailored to the local language and context to systematically detect, flag, and respond to this form of violence. This project intervenes by creating a technology-driven tool, accompanied by capacity building and advocacy, to fill that gap and support safer, more respectful digital discourse.
Objectives
- Platform Development & Piloting: Build and pilot an AI tool capable of identifying hate speech against women with high accuracy (≈85%) in Armenian, Latin, Cyrillic, and transliterated scripts.
- Capacity Building: Train 20 media professionals and 20 women public figures / influencers in understanding, using, and responding to hate speech cases.
- Awareness Raising & Demand Creation: Conduct public campaigns and informational outreach to promote adoption and understanding of the tool among media, CSOs, and the broader public.
- Institutionalization & Advocacy: Engage with Media Ethics Council, media outlets, and other stakeholders to integrate the tool into self-regulation processes and policy frameworks.
- Sustainability & Uptake: Establish a viable subscription or licensing model post-project, and ensure the technical sustainability of the tool via dedicated support structures.
———–
This project is funded with UK International Development from the UK government and is implemented by the All for Equal Rights Foundation within the framework of the “Building Resilience in the Eastern Neighbourhood” program, carried out by the Institute for War & Peace Reporting (IWPR)
