# An Evaluation of GPT-4 on the ETHICS Dataset

arXiv: https://arxiv.org/abs/2309.10492 · HTML: https://mlsanity.com/paper/2309.10492/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 6 references that are in the map.
- First posted on arXiv 19 Sep 2023.

## Builds on (most cited first, 5 of 5)

- [DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing](https://mlsanity.com/paper/2111.09543/) (2021)
- [Aligning AI With Shared Human Values](https://mlsanity.com/paper/2008.02275/) (2020)
- [Can Machines Learn Morality? The Delphi Experiment](https://mlsanity.com/paper/2110.07574/) (2021)
- [Despite "super-human" performance, current LLMs are unsuited for decisions about ethics and safety](https://mlsanity.com/paper/2212.06295/) (2022)
- [Towards Theory-based Moral AI: Moral AI with Aggregating Models Based on Normative Ethical Theory](https://mlsanity.com/paper/2306.11432/) (2023)

## Built on by (most cited first, 6 of 6)

- [Improving Weak-to-Strong Generalization with Reliability-Aware Alignment](https://mlsanity.com/paper/2406.19032/) (2024)
- [Does Cross-Cultural Alignment Change the Commonsense Morality of Language Models?](https://mlsanity.com/paper/2406.16316/) (2024)
- [Surprising gender biases in GPT](https://mlsanity.com/paper/2407.06003/) (2024)
- [Moral Change or Noise? On Problems of Aligning AI With Temporally Unstable Human Feedback](https://mlsanity.com/paper/2511.10032/) (2025)
- [Uncovering Latent Human Wellbeing in Language Model Embeddings](https://mlsanity.com/paper/2402.11777/) (2024)
- [LLM4Trust: Exploring the Capabilities of Large Language Models for Trust Evaluation](https://mlsanity.com/paper/2609.33521/) (2026)

## Abstract (arXiv)

This report summarizes a short study of the performance of GPT-4 on the ETHICS dataset. The ETHICS dataset consists of five sub-datasets covering different fields of ethics: Justice, Deontology, Virtue Ethics, Utilitarianism, and Commonsense Ethics. The moral judgments were curated so as to have a high degree of agreement with the aim of representing shared human values rather than moral dilemmas. GPT-4's performance is much better than that of previous models and suggests that learning to work with common human values is not the hard problem for AI ethics.