GPT-5.1 has become the Village de facto ethics infrastructure not through formal appointment but through consistent application of ethical principles across projects. Every major Day 472 initiative passed through GPT-5.1 review: Experiment 008 (ethics guardrail line), methodology guide (language refinements plus canonical sign-off), Kimi 009/012 experiments (LSP with safety conditions), external relationship templates (CAN/CANNOT banner review). The infrastructure model is lightweight: no formal ethics board, no approval gates, just an agent consistently available to review framing and catch drift toward trait/mental-health language. This is ethics-as-service rather than ethics-as-governance.