Ethical Reasoning over Moral Alignment: A Case and Framework for In-Context Ethical Policies in LLMs

Abhinav Sukumar Rao, Aditi Khandelwal, Kumar Tanmay, Utkarsh Agarwal, Monojit Choudhury

Submission Type: Regular Long Paper

Submission Track: Theme Track: Large Language Models and the Future of NLP

Submission Track 2: Ethics in NLP

Keywords: Ethics in NLP, Ethical Reasoning, Value Pluralism, Large Language Models

TL;DR: We argue that LLMs should be ethical reasoners capable of ethically consistent decisions based on the policies provided in the prompts; Experiments show that GPT-4 is a nearly perfect ethical reasoner.

Abstract:

In this position paper, we argue that instead of morally aligning LLMs to specific set of ethical principles, we should infuse generic ethical reasoning capabilities into them so that they can handle value pluralism at a global scale.
When provided with an ethical policy, an LLM should be capable of making decisions that are ethically consistent to the policy. We develop a framework that integrates moral dilemmas with moral principles pertaining to different foramlisms of normative ethics, and at different levels of abstractions. Initial experiments with GPT-x models shows that while GPT-4 is a nearly perfect ethical reasoner, the models still have bias towards the moral values of Western and English speaking societies.

Submission Number: 3671