Search site...

Developing "Code Caché": A Collaborative Approach to Addressing Bias in AI

Artificial Intelligence systems are not neutral; they reflect the data and decisions of the societies that create them. When left unchecked, algorithms can inadvertently reproduce and amplify existing social inequalities. To address this challenge constructively, the European project DIVERSIFAIR partnered with La Fresque du Sexisme to develop “Code Caché” (The Hidden Code), an educational, collaborative workshop designed to make the complex topic of algorithmic bias accessible to all.

Here is an overview of how this partnership was formed, the insights gained during its pilot phase, and its integration into our permanent educational resources.

Bridging Research and Public Awareness

The primary goal of this collaboration was to translate academic research on intersectional AI bias into a practical, engaging learning experience. The initiative brought together a multidisciplinary team to ensure both pedagogical accessibility and technical accuracy: Anabelle Pasillas (Founder of Egaliki, Gender Equality Consultant, Corporate Trainer, and Keynote Speaker), Roxane de Pelet (Founder of Projet Epicène and Co-founder of La Fresque du Sexisme) and Gabriela Del Barco (DEI Consultant, Facilitator, and Responsible & Equitable AI Specialist).

This core team worked in close coordination with representatives from the DIVERSIFAIR consortium, including Steven Vethman (Sciences Po) and Louise Aupetit (Women in AI), to align the workshop with the project’s European research framework.

Insights from the Pilot Phase: 9 Workshops, 100+ Participants

To refine the tool, the workshop was evaluated through 9 pilot sessions involving 100+ participants across various sectors. These sessions allowed us to test the content with diverse audiences and gather constructive feedback, including

  • Technical and Academic Sectors: At events like the Assises du Genre et du Numérique in Switzerland, the Salon DATA IA, and the Semaine de l’Entreprise Responsable et Inclusive (SERI), the workshop engaged data scientists and developers. It allowed tech professionals to reflect on the societal implications of their work, with one participant noting: “I appreciated questioning the impacts of our decisions on the AI we create and use, but also the impacts of this technology on our society. The workshop integrates several real cases.”
  • Corporate and Civil Society Groups: Sessions with communication, HR, and marketing teams at Concentrix, as well as advocates from the Sista collective and Ouvre Boîtes, explored how algorithmic design impacts professional inclusion and organisational communication.
  • Students and Future Professionals: Workshops held for interns at Euroclear France and digital marketing students at ISEG Nantes highlighted the value of early digital literacy. As one student shared: “I better understand the steps behind the development of an AI now, and I realise there is a hidden side to it.”

Addressing Barriers to Digital Literacy

In line with DIVERSIFAIR’s commitment to inclusion, the pilot phase specifically aimed to lower barriers to understanding technology. An intersectional analysis of our 110 participants shows that the sessions successfully reached:

  • 86% participants facing social barriers (including women, non-binary individuals, and younger participants who are statistically underrepresented in tech design).
  • 92% participants without technical backgrounds, proving that algorithmic literacy can be successfully taught to non-experts.
  • 31% participants facing economic barriers (students and young interns), emphasizing the importance of providing free access to foundational digital education.

The pilot phase also provided valuable constructive feedback: sessions with younger audiences showed us where to further simplify complex concepts, while feedback from tech experts helped us make the technical case studies even more precise.

Next Steps and Future Availability

The pilot phase demonstrated that when citizens and professionals are given the right tools, they can move from passive users of AI to active stakeholders in digital ethics.

As the DIVERSIFAIR project concludes its initial European funding cycle, we are pleased to announce that “Code Caché” is now a permanent fixture in La Fresque du Sexisme’s catalogue. This ensures that the research and tools developed over the last three years will continue to serve as a lasting resource for companies, schools, and civic organisations seeking to foster a more responsible digital future.

Interested in organising a workshop? You can find more information and book a session directly through La Fresque du Sexisme website.

Please note: To ensure maximum impact within the local communities where it was developed, the workshop is currently conducted exclusively in French.

Project

Goals

Consortium

hey

News & Events

Resources

Results

hey

Privacy Policy

Contact

hey

 

Stay in the Loop!

Don’t miss out on the latest updates, subscribe to our newsletter and be the first to know what’s happening.

Subscribe

This project has received funding from the European Education and Culture Executive Agency (EACEA) in the framework of Erasmus+, EU solidarity Corps A.2 – Skills and Innovation under grant agreement 101107969.

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.