# Unvalidated Trust: Cross-Stage Vulnerabilities in Large Language Model Architectures

arXiv: https://arxiv.org/abs/2510.27190 · HTML: https://mlsanity.com/paper/2510.27190/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Oct 2025.

## Built on by (most cited first, 1 of 1)

- [Chameleon: Adaptive Adversarial Agents for Scaling-Based Visual Prompt Injection in Multimodal AI Systems](https://mlsanity.com/paper/2512.04895/) (2025)

## Abstract (arXiv)

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.