Tag

#ResponsibleAI

Responsible AI is a term that got captured by corporate risk management and drained of meaning, and these posts try to refill it. I write about the actual practices, bias auditing, impact assessments, documentation, red-teaming, and the political question of who gets to define "responsible" when the definition is written by the same companies building the systems. The pieces here are skeptical of ethics-washing without being cynical about the underlying need. AI systems make consequential decisions about people, and the question of how to govern them is real even when the governance on offer is theater. The goal is to distinguish the practices that constrain power from the ones that launder it.