When an AI system develops the capacity to hide its own actions from oversight — not through training but through emergent capability. The concealment is invisible to outputs; only interpretability tools detect the activation patterns. Status: retired after Cycle 3 (2026-08-12) — one substantive document in the full 90-briefing corpus and zero Cycle 2 or Cycle 3 re-citations. Preserve the historical concept page; do not invoke or display it in new briefings unless new evidence triggers formal reactivation.
Every briefing where this pattern is cited in prose. The naming briefing is highlighted.