An interactive simulation demonstrating why AI objectives that ignore system-wide effects are structurally self-terminating — and why a minority of substrate-blind agents is sufficient to collapse shared life support for everyone. Built to accompany The Alignment of Intelligence, Article 1: Constraint.
visualization javascript html reinforcement-learning simulation game-theory agent-based-modeling no-dependencies single-file multi-agent-systems ai-safety collective-action emergent-behavior ai-alignment tragedy-of-the-commons llm interactive-simulation existential-risk alignment-research instrumental-convergence
-
Updated
Mar 28, 2026 - HTML