Event Details
SMUCDL was delighted to host our External Research Fellow Assoc Prof Gilad Abiri from Peking University School of Transnational Law for an insightful seminar on Epistemic Subordination.
Based on a new piece co-authored with Prof Emanuel V. Towfigh, the seminar examined how generative AI may encode the majority’s way of knowing as the default infrastructure of knowledge itself. Assoc Prof Abiri explored how minority epistemologies may remain present in training data, yet become structurally subordinated in model outputs through the statistical baselines and dominant cultural frameworks embedded in large-scale AI systems.
The discussion considered why this form of harm may not be adequately addressed through conventional bias audits or downstream regulation. Instead, the paper argues that if epistemic subordination is produced at the level of model training, legal and governance responses must also learn to engage at that level.
We are grateful to Assoc Prof Abiri for sharing this thought-provoking work and for advancing important conversations on generative AI, knowledge production, equality, and technology governance.
