The Moral Consistency Pipeline: Continuous Ethical Evaluation for Large Language Models

The rapid advancement and adaptability of Large Language Models (LLMs) highlight the need for moral consistency, the capacity to maintain ethically coherent reasoning across varied contexts. Existing alignment frameworks, structured approaches designed to align model behavior with human ethical and social norms, often rely on static datasets and post-hoc evaluations, offering limited insight into how ethical reasoning may evolve across different contexts and temporal scales. This study presents the Moral Consistency Pipeline (MoCoP), a dataset-free, closed-loop framework for continuously evaluating and interpreting LLMs’ moral stability. MoCoP combines three supporting layers: 1) lexical integrity analysis, 2) semantic risk estimation, and 3) reasoning-based judgment modeling within a self-sustaining architecture that autonomously generates, evaluates, and refines ethical scenarios without external supervision. Our empirical results on GPT-4-Turbo and DeepSeek suggest that MoCoP effectively captures longitudinal ethical behavior, revealing a strong inverse relationship between ethical and toxicity dimensions (rET = −0.81, p < 0.001) and a near-zero association with response latency (rEL ≈ 0). These findings demonstrate that moral coherence and linguistic safety tend to emerge as stable, interpretable characteristics of model behavior rather than as short-term fluctuations. Furthermore, by reframing ethical evaluation as a dynamic, model-agnostic form of moral introspection, MoCoP offers a reproducible foundation for scalable, continuous auditing and advances the study of computational morality in autonomous AI systems.