INTERVISIONS

INTERVISIONS is a multidisciplinary project combining intersectional feminist theory, participatory AI, and deep learning research to build participatory bias audit tools and fairness benchmarks for vision–language foundation models.
We co-create methods and datasets with communities and develop Intersectional Impact Assessment (IIA) guidelines to promote responsible and inclusive AI development.

  • Focus: Participatory bias auditing, fairness benchmarking, and intersectional impact assessment for vision–language models
  • Funding: EU CERV programme (Call: CERV-2024-CHAR-LITI-CHARTER) — Project ID: 101214711
  • Participants: Alia/Donestech, Computer Vision Center (CVC), Diputació de Barcelona (DIBA)
Funded by the European Union
Funded by the European Union under the CERV programme (Call: CERV-2024-CHAR-LITI-CHARTER, Project ID: 101214711). The content reflects only the authors’ views; the European Union is not responsible for any use that may be made of the information it contains.