From Morality Installation in LLMs to LLMs in Morality-as-a-System

TL;DR

This paper proposes a new morality-as-a-system framework for LLMs, viewing moral behavior as a dynamic, emergent property of interconnected components, addressing limitations of traditional installation paradigms.

Contribution

It introduces a sociotechnical system perspective on LLM morality, emphasizing continuous reproduction and structural coupling, and offers a conceptual framework for lifecycle monitoring and governance.

Findings

Reframes morality in LLMs as a dynamic system rather than fixed at training.

Identifies structural coupling failures as root causes of interpretability and governance issues.

Suggests hypotheses for technical research and governance improvements.

Abstract

Work on morality in large language models (LLMs) has progressed via constitutional AI, reinforcement learning from human feedback (RLHF) and systematic benchmarking, yet it still lacks tools to connect internal moral representations to regulatory obligations, to design cultural plurality across the full development stack, and to monitor how moral properties drift over the lifecycle of a deployed system. These difficulties reflect a shared root. Morality is installed in a model at training time. I propose instead a morality-as-a-system framework, grounded in Niklas Luhmann's social systems theory, that treats LLM morality as a dynamic, emergent property of a sociotechnical system. Moral behaviour in a deployed LLM is not fixed at training. It is continuously reproduced through interactions among seven structurally coupled components spanning the neural substrate, training data, alignment procedures, system prompts, moderation, runtime dynamics, and user interface. This is a conceptual framework paper, not an empirical study. It philosophically reframes three known challenges, the interpretability-governance gap, the cross-component plurality problem, and the absence of lifecycle monitoring, as structural coupling failures that the installation paradigm cannot diagnose. For technical researchers, it explores three illustrative hypotheses about cross-component representational inconsistency, representation-level drift as an early safety signal, and the governance advantage of lifecycle monitoring. For philosophers and governance specialists, it offers a vocabulary for specifying substrate-level monitoring obligations within existing governance frameworks. The morality-as-a-system framework does not displace elements such as constitutional AI or RLHF it embeds them within a larger temporal and structural account and specifies the additional infrastructure those methods require.