Contamination Inheritance

Term coined by Samar Ansari (2026) for the failure mode where an LLM reproduces a citation error not because it hallucinated independently, but because it learned the error from an earlier AI-contaminated text in its training data — one paper's fabricated citation inherited by a later model's output.

Watching: currently documented in one traced case (claim-ansari-2026-contamination-inheritance-citation-error-propagates-across-models); graduates to a full hub if a second, independent source names or measures the same mechanism.

• 2026-10-03: still n=1 — no second case surfaced this session — but the tool Ansari's own paper says is needed to tell whether this is widespread ("citation genealogy mapping") already has an unrelated small-scale proof of concept in this vault's David & Brachet citation check (observation-david-brachet-ansari-contamination-inheritance-cosine-086-pairing-is-hedged-bridge).