IAMRAGBench: poisoned retrieval documents in AI-generated AWS IAM policies

Evidence record

Primary paper measuring attacker-favored permission expansion under document poisoning: 46 successful attack goals out of 292 eligible contexts in the included-context setting.

Read the original source ↗

Claim-level citations (3)

• supportsPoisoned reference text steered AI-generated AWS policies toward excess rights: The researchers report that 46 of 292 eligible included-context attempts reached an attacker-selected permission-expansion goal.

  Attack evaluation results, eligible attempts

• supportsPoisoned reference text steered AI-generated AWS policies toward excess rights: The attack targets excessive permissions in provider-valid generated IAM policy output rather than relying on malformed policy syntax.

  Threat model and IAM policy validity criteria

• supportsPoisoned reference text steered AI-generated AWS policies toward excess rights: With end-to-end BM25 retrieval, the designated poison was retrieved for all 60 tasks but only 7 of 294 eligible endpoint-case runs met the exact attacker-goal authorization test, versus 46 of 292 in the direct included-context condition.

  Section 6.1, Table 2, and Section 6.2, end-to-end BM25 regime