{"ID":2847965,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.27190","arxiv_id":"2510.27190","title":"Unvalidated Trust: Cross-Stage Vulnerabilities in Large Language Model Architectures","abstract":"As Large Language Models (LLMs) are increasingly integrated into automated, multi-stage pipelines, risk patterns that arise from unvalidated trust between processing stages become a practical concern. This paper presents a mechanism-centered taxonomy of 41 recurring risk patterns in commercial LLMs. The analysis shows that inputs are often interpreted non-neutrally and can trigger implementation-shaped responses or unintended state changes even without explicit commands. We argue that these behaviors constitute architectural failure modes and that string-level filtering alone is insufficient. To mitigate such cross-stage vulnerabilities, we recommend zero-trust architectural principles, including provenance enforcement, context sealing, and plan revalidation, and we introduce \"Countermind\" as a conceptual blueprint for implementing these defenses.","short_abstract":"As Large Language Models (LLMs) are increasingly integrated into automated, multi-stage pipelines, risk patterns that arise from unvalidated trust between processing stages become a practical concern. This paper presents a mechanism-centered taxonomy of 41 recurring risk patterns in commercial LLMs. The analysis shows...","url_abs":"https://arxiv.org/abs/2510.27190","url_pdf":"https://arxiv.org/pdf/2510.27190v1","authors":"[\"Dominik Schwarz\"]","published":"2025-10-30T09:38:45Z","proceeding":"cs.CR","tasks":"[\"cs.CR\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}