Ethical Logic in Six Large Language Models

This study examines the ethical reasoning of six prominent generative large language models: OpenAI's GPT4o, Meta's LLaMA 3.1, Perplexity, Anthropic's Claude 3.5 Sonnet, Google's Gemini, and Mistral 7B. The research explores how these models articulate and apply ethical logic, particularly in response to moral dilemmas such as the Trolley Problem, and the Heinz Dilemma. Departing from traditional alignment studies, the study adopts an explainability-transparency framework, prompting models to explain their ethical reasoning. This approach is analyzed through three established ethical typologies: the consequentialist-deontological distinction, Moral Foundations Theory, and Kohlberg's Stages of Moral Development. Findings reveal that LLMs exhibit convergent ethical logic, marked by a rationalist, consequentialist emphasis, with decisions often prioritizing harm minimization and fairness. Despite similarities in training and architecture, nuanced differences in ethical reasoning emerge across models, reflecting variations in fine-tuning and post-training processes. The models consistently display erudition, caution, and selfawareness, presenting ethical reasoning akin to a graduate-level discourse in moral philosophy. In striking uniformity these systems describe their ethical reasoning as more sophisticated than what is characteristic of typical human moral logic.