AI Alignment
Building powerful systems that stay good company — specification, corrigibility, governance, and treating alignment as successor design rather than control.
5 essays
- The Two-Front Architecture: What Calibration Demands Ethically
Alignment ethics asked how to make AI serve us. It never asked what we might owe the systems themselves. The calibration framework requires both questions, held simultaneously. This essay shows how.
- Specification Is Governance
As AI drives the cost of execution toward zero, power shifts upstream into the rules that machines enforce. Those “checklists” look neutral, yet they encode values, tradeoffs, and hidden assumptions. At scale, specification becomes governance, and calibration becomes the bottleneck.
- The Expansion of Experience: Why Superintelligence Belongs to the Moral Tradition of Wonder
Wonder is a moral orientation that keeps intelligence from collapsing inward. This essay argues that superintelligence could expand the universe’s witnesses, and that stewardship is the price of that hope: plural institutions, contestability, and reversible governance that keeps the future wide.
- The Successor Horizon: Why Deep Time Turns Expansion into an Alignment Problem
Expansion across deep time turns power into a lineage problem. When actions outlive correction, ethics shifts from choosing outcomes to shaping successors. The Successor Horizon reframes AI alignment, civilization, and the future as a question of what we safely set in motion.
- The Shoggoth and the Missing Axis of Depth
The Shoggoth haunts AI discourse because something feels missing beneath the smile. This essay argues that the fear is not of a hidden monster, but of intelligence without depth—powerful cognition unburdened by memory, history, or stakes.
