Embedding EDI and RRI in AI Governance: Lessons from a Collaborative Toolkit Project
by Peter Winter and John Downer (University of Bristol), Joshua Krook and Jan Blockx (University of Antwerp)
Our project, “Transparency Regulation Toolkits for Responsible AI,” a collaboration between the University of Bristol and the University of Antwerp, aimed to build practical tools for AI governance – and to embed Responsible Research and Innovation (RRI) and Equality, Diversity, and Inclusion (EDI) into that process.
Our project team wasn’t in itself diverse, and we recognised this as a limitation. To compensate, we focused on building EDI and RRI into the way we engaged with stakeholders and developed the toolkit.
We held two stakeholder workshops – one in Bristol, the other in Antwerp – with deliberately varied stakeholders: academics, SME representatives, AI ethics consultants, policymakers, and professionals from large enterprises. Each workshop followed the Responsible Innovation framework, with sessions focused on reflexivity (“What are you currently doing to be transparent about AI?”) and anticipation (“How can we do transparency better?”).
To support equitable discussion, we did everything we could to ensure gender-balanced breakout groups. While we couldn’t control last-minute attendance, we sent reminders and follow-up messages to maintain fairness and balance.
We didn’t look for checklist answers. Instead, we used facilitated discussion to surface grounded insights – what’s working, what’s missing, and what support is needed. Feedback, especially from those outside technical AI spaces, led us to revise the toolkit: using plainer language, clarifying assumptions, and reframing use cases.
The main lesson? You don’t need to get everything right from the start – but you do need to take it seriously. RRI and EDI aren’t abstract ideals. They’re practical tools that, when used with care, make the research better.