Adversarial Moral Stress Testing of Large Language Models

Evaluating the ethical robustness of large language models (LLMs) deployed in software systems remains challenging, particularly under sustained adversarial user interaction. This paper introduces Adversarial Moral Stress Testing (AMST), a stress-based evaluation framework for assessing ethical robustness under adversarial multi-round interactions. AMST applies structured stress transformations to prompts and evaluates model behavior through distribution-aware robustness metrics. Results demonstrate substantial differences in robustness profiles across models and expose degradation patterns not observable under conventional single-round evaluation.