About DataBias
DataBias connects researchers, funders, and anyone working in AI with the women of colour doing some of the most important work in the field — who are still, somehow, the hardest to find.
Ask most people to name five women of colour working on AI safety or AI governance and they'll struggle past two or three — usually the same two or three. Not because the work doesn't exist, but because the infrastructure to surface it doesn't.
The researchers are there. The policy analysts, ethicists, governance specialists, and dataset auditors are there. They're publishing, teaching, building, testifying. They're just harder to find — particularly if you're not already in the right networks, or based outside the US.
DataBias is a free, public, searchable directory of women of colour working in AI safety, AI governance, and algorithmic justice. It exists so that researchers can be found by funders, conference organisers can find speakers, journalists can find sources, and anyone entering this field can see clearly who is already doing the work.
DataBias covers three overlapping areas:
Most AI safety and governance discourse is dominated by voices from the United States and, to a lesser extent, the UK. That reflects funding patterns more than the distribution of important work.
DataBias actively seeks out researchers based in Africa, South and Southeast Asia, Latin America, the Caribbean, the Middle East, and other regions consistently underrepresented in mainstream AI conversations. The database is designed to counterbalance the default, not reproduce it.
DataBias operates like a professional research directory: we list publicly available information about researchers whose work is relevant to the field. All information comes from public sources — institutional pages, published papers, official bios, public social profiles.
Every profile has a "claim or update" link. If a researcher wants to write their own bio, update their information, add links to recent work, or be removed entirely — that is straightforward and respected immediately.
We also reach out directly to researchers to invite them to shape how they're represented. The goal is always for a profile to reflect what the researcher wants the world to know about their work — not just what we could find about them.
DataBias is guided by the Black Women Best Framework (Janelle Jones, Roosevelt Institute, 2020): when we design systems for those navigating the most structural barriers, we build infrastructure that works better for everyone. It's not a political statement. It's a design principle backed by evidence — more diverse research communities produce more robust findings, catch more failure modes, and build more broadly useful tools.
If you're a woman of colour working in AI safety, governance, or algorithmic justice — anywhere in the world — get in touch or use the submission link on the catalogue page. If you want to nominate someone, suggest a resource, or explore a partnership, we'd love to hear from you.
Join the Directory
Submit your profile — it's free, takes 10 minutes, and puts your work in front of people actively looking for it.
Submit Your Profile →"When we design for those navigating the most structural barriers, we build infrastructure that works for everyone." — Black Women Best Framework, Janelle Jones (2020)