Empowering Communities to Audit AI: The New Community-Led AI Audit Guide Scrum Handbook

AI systems affect our daily lives in ways that are often invisible, but their impacts can be profound.

The DIVERSIFAIR project is proud to release the Community-Led AI Audit Guide, a practical

toolkit designed to help communities, civil-society organisations, and independent auditors

investigate AI’s real-world effects.

Developed by Eticas within the DIVERSIFAIR project, this guide equips users with:

• A step-by-step auditing process for assessing AI systems without insider access.

• A socio-technical approach combining technical methods (scraping, bot testing, comparative audits) with qualitative research (interviews, lived experience analysis).

• Seven adaptable audit methods suitable for AI systems like recommender algorithms, facial recognition, and pricing tools.

• Case studies demonstrating how community-led audits have exposed bias and discrimination across Europe.

Whether you are a civil society advocate, journalist, researcher, or data scientist, this guide makes accountable AI oversight accessible to all.

📥 Read the guide and start auditing AI in your community: https://diversifair-project.eu/courses/community-led-ai-audit-guide/

ABOUT DIVERSIFAIR

DIVERSIFAIR is an Erasmus+ project that aims to foster intersectional fairness in AI systems by creating educational resources, research, and policy guidance. Women in AI is proud to be a partner in this initiative, working alongside leading academic institutions, AI educators, and civil society organisations to build a more inclusive and ethical future for AI.

Find out more about DIVERSIFAIR.

Stay in touch

Follow DIVERSIFAIR on LinkedIn and register to the project's newsletter.

Funded by the European Union. Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of the European Union or the Culture Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.