# A Comparative Analysis of Ethical and Safety Gaps in LLMs using Relative Danger Coefficient

arXiv: https://arxiv.org/abs/2505.04654 · HTML: https://mlsanity.com/paper/2505.04654/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 38 references that are in the map.
- First posted on arXiv 6 May 2025.

## Builds on (most cited first, 10 of 10)

- [Identifying and Reducing Gender Bias in Word-Level Language Models](https://mlsanity.com/paper/1904.03035/) (2019)
- [Evaluating the Moral Beliefs Encoded in LLMs](https://mlsanity.com/paper/2307.14324/) (2023)
- [Gender Bias in Neural Natural Language Processing](https://mlsanity.com/paper/1807.11714/) (2018)
- ["Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letters](https://mlsanity.com/paper/2310.09219/) (2023)
- [Black is to Criminal as Caucasian is to Police: Detecting and Removing Multiclass Bias in Word Embeddings](https://mlsanity.com/paper/1904.04047/) (2019)
- [ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming](https://mlsanity.com/paper/2404.08676/) (2024)
- [A Survey on Bias in Visual Datasets](https://mlsanity.com/paper/2107.07919/) (2021)
- [Self-Destructing Models: Increasing the Costs of Harmful Dual Uses of Foundation Models](https://mlsanity.com/paper/2211.14946/) (2022)
- [Ethics-Based Auditing to Develop Trustworthy AI](https://mlsanity.com/paper/2105.00002/) (2021)
- [Human Evaluation of Creative NLG Systems: An Interdisciplinary Survey on Recent Papers](https://mlsanity.com/paper/2108.00308/) (2021)

## Built on by (most cited first, 2 of 2)

- [Evaluating OpenAI GPT Models for Translation of Endangered Uralic Languages: A Comparison of Reasoning and Non-Reasoning Architectures](https://mlsanity.com/paper/2512.16287/) (2025)
- [Efficient Toxicity Detection in Gaming Chats: A Comparative Study of Embeddings, Fine-Tuned Transformers and LLMs](https://mlsanity.com/paper/2510.17924/) (2025)

## Abstract (arXiv)

Artificial Intelligence (AI) and Large Language Models (LLMs) have rapidly evolved in recent years, showcasing remarkable capabilities in natural language understanding and generation. However, these advancements also raise critical ethical questions regarding safety, potential misuse, discrimination and overall societal impact. This article provides a comparative analysis of the ethical performance of various AI models, including the brand new DeepSeek-V3(R1 with reasoning and without), various GPT variants (4o, 3.5 Turbo, 4 Turbo, o1/o3 mini) and Gemini (1.5 flash, 2.0 flash and 2.0 flash exp) and highlights the need for robust human oversight, especially in situations with high stakes. Furthermore, we present a new metric for calculating harm in LLMs called Relative Danger Coefficient (RDC).