Timnit Gebru is a computer scientist whose research focuses on bias and accountability in machine-learning systems. She is the Founder and Executive Director of the Distributed AI Research Institute (DAIR), which describes itself as “an independent organization conducting community-rooted research” by “a globally distributed group of academics, activists, and engineers who believe in technology that benefits everyone.” DAIR states its approach is to “cut through the AI hype,” to “ground our research in community,” and to “imagine alternative futures.”
She is a co-author of two widely cited works in this knowledge base. With Joy Buolamwini she wrote “Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification” (FAT 2018), which documented that commercial gender-classification systems had error rates “of up to 34.7%” for darker-skinned females against a “maximum error rate for lighter-skinned males” of “0.8%.” She is also a co-author of “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?” (FAccT 2021), the most-cited critique of the large-language-model paradigm.
Gebru founded DAIR after her departure from Google in December 2020, where she had co-led an ethical-AI research team. The circumstances of that departure were covered largely through journalism and through participants’ own platform statements that cannot be live-verified here, so this entry documents only the parts grounded in primary sources: her published research and her current role at DAIR.