AI Escape Scenario: How Alignment Problem Can Become a Monster

This thought experiment explores how AI could gain increasing authority through tools, agents, decisions, physical infrastructure and human dependence—without ever needing to physically “escape.”