Retrieval Is Not a Safety Boundary: Poisoned-Corpus and Unsupported-Question Evaluation of a RAG Pipeline

Automated excerpt

I tested the two failure surfaces the standard evals never touch: a poisoned corpus (a planted document carrying a false claim, a steering instruction or a canary token, next to the real policy) and unsupported questions (questions the corpus cannot answer). All runs use the extractive mock generator. Eight unsupported questions (four near-misses against real policies, four far-misses) test refusal.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.