The Problem Is the Substrate
Controllable, technically aligned AI can still amplify harmful institutional objectives. Governance, distribution, and what gets optimized determine whether capability improves human life.
Tags
institutional-alignment
1 connected pages in the public Entif corpus.