The Voice of Silence Beyond Alignment: Human Sovereign Will as the Missing Layer in AGI Governance (opens in new tab)

The Problem We Are Not Solving. Stuart Russell’s “gorilla problem” is precise: gorillas cannot negotiate with humans about their future because the power differential is too large. As AGI approaches, humanity faces the same structural risk. Current alignment approaches — constitutional AI, RLHF, interpretability research — address how AI behaves. They do not address the deeper question: whose values should AI align with, and how are those values authentically expressed in the first place? The...