Sabelo Mhlambi*

About

I work on AI safety, competition, and model evaluation — and build the models I write about.

At Bhala AI, which I founded, I pretrain speech and language models from scratch for communities AI has excluded. I audit open frontier models for the harms and bias their representations carry. I wrote the antitrust analysis for the U.S. House Judiciary Committee’s Subcommittee on Antitrust, Commercial, and Administrative Law as a Senior Congressional Innovation Fellow.

My research background is in AI ethics and human rights. From Rationality to Relationality, published by the Carr Center for Human Rights Policy at Harvard Kennedy School, argues that the Western grounding of personhood in rationality is formally incomplete, and that ubuntu’s relational personhood offers a firmer foundation for AI governance. I founded Bantucracy, a research initiative on ubuntu ethics and AI.

Currently

Founder, Bhala AI
AI Ethics Advisory Board, The Institute for Responsible AI
Associate Editor, Human Rights and Global Perspectives, Critical AI

Previously

U.S. House Judiciary Committee, Subcommittee on Antitrust, Commercial, and Administrative Law — Senior Congressional Innovation Fellow, TechCongress
Carr Center for Human Rights Policy, Harvard Kennedy School
Berkman Klein Center for Internet & Society, Harvard University
Digital Civil Society Lab, Stanford University
Software engineering at Mapbox and Iris.tv

* Audio generated by Bhala AI.

This personal website expresses the opinions of none of these organizations.