From Benchmarks to Behaviors: A Framework for Auditing AI Agents
Source check: Source checked. Not independent scientific confirmation.
Finding / tested result: A controlled fixed-source study reports that a user-memory condition changed agents’ citation choices, including a measured increase in misinformation or conspiracy citations.
Technology: Controlled audit of AI agents using fixed sources and user-memory conditions
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Controlled audit of AI agents using fixed sources and user-memory conditions is used in the reported analysis or evaluation.
- Limitation / still unknown
- The citation-choice change, including misinformation and conspiracy citations, is limited to the controlled source set and study conditions.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://arxiv.org/html/2611.03939v1
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.