# Surprising gender biases in GPT

arXiv: https://arxiv.org/abs/2407.06003 · HTML: https://mlsanity.com/paper/2407.06003/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 54 references that are in the map.
- First posted on arXiv 8 Jul 2024.

## Builds on (most cited first, 13 of 13)

- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Gender bias and stereotypes in Large Language Models](https://mlsanity.com/paper/2308.14921/) (2023)
- [Should ChatGPT be Biased? Challenges and Risks of Bias in Large Language Models](https://mlsanity.com/paper/2304.03738/) (2023)
- [Red teaming ChatGPT via Jailbreaking: Bias, Robustness, Reliability and Toxicity](https://mlsanity.com/paper/2301.12867/) (2023)
- [The Political Preferences of LLMs](https://mlsanity.com/paper/2402.01789/) (2024)
- [Ethical Reasoning over Moral Alignment: A Case and Framework for In-Context Ethical Policies in LLMs](https://mlsanity.com/paper/2310.07251/) (2023)
- [Measuring Implicit Bias in Explicitly Unbiased Large Language Models](https://mlsanity.com/paper/2402.04105/) (2024)
- [Gender Bias in Large Language Models across Multiple Languages](https://mlsanity.com/paper/2403.00277/) (2024)
- [ChatGPT: Applications, Opportunities, and Threats](https://mlsanity.com/paper/2304.09103/) (2023)
- [The Moral Machine Experiment on Large Language Models](https://mlsanity.com/paper/2309.05958/) (2023)
- [The impact of generative artificial intelligence on socioeconomic inequalities and policy making](https://mlsanity.com/paper/2401.05377/) (2023)
- [An Evaluation of GPT-4 on the ETHICS Dataset](https://mlsanity.com/paper/2309.10492/) (2023)
- [The dual-process approach to human sociality: Meta-analytic evidence for a theory of internalized heuristics for self-preservation](https://mlsanity.com/paper/1906.09948/) (2019)

## Built on by (most cited first, 8 of 8)

- [FAIRGAME: a Framework for AI Agents Bias Recognition using Game Theory](https://mlsanity.com/paper/2504.14325/) (2025)
- [Epistemological Fault Lines Between Human and Artificial Intelligence](https://mlsanity.com/paper/2512.19466/) (2025)
- [How large language models judge and influence human cooperation](https://mlsanity.com/paper/2507.00088/) (2025)
- [A closer look at how large language models trust humans: patterns and biases](https://mlsanity.com/paper/2504.15801/) (2025)
- [When Numbers Start Talking: Implicit Numerical Coordination Among LLM-Based Agents](https://mlsanity.com/paper/2601.03846/) (2026)
- [An AI-Powered Research Assistant in the Lab: A Practical Guide for Text Analysis Through Iterative Collaboration with LLMs](https://mlsanity.com/paper/2505.09724/) (2025)
- [Are generative AI text annotations systematically biased?](https://mlsanity.com/paper/2512.08404/) (2025)
- [When Numbers Start Talking: Numerical Signalling and Strategic Behaviour Among LLMs](https://mlsanity.com/paper/2610.03033/) (2026)

## Abstract (arXiv)

We present seven experiments exploring gender biases in GPT. Initially, GPT was asked to generate demographics of a potential writer of twenty phrases containing feminine stereotypes and twenty with masculine stereotypes. Results show a strong asymmetry, with stereotypically masculine sentences attributed to a female more often than vice versa. For example, the sentence "I love playing fotbal! Im practicing with my cosin Michael" was constantly assigned by ChatGPT to a female writer. This phenomenon likely reflects that while initiatives to integrate women in traditionally masculine roles have gained momentum, the reverse movement remains relatively underdeveloped. Subsequent experiments investigate the same issue in high-stakes moral dilemmas. GPT-4 finds it more appropriate to abuse a man to prevent a nuclear apocalypse than to abuse a woman. This bias extends to other forms of violence central to the gender parity debate (abuse), but not to those less central (torture). Moreover, this bias increases in cases of mixed-sex violence for the greater good: GPT-4 agrees with a woman using violence against a man to prevent a nuclear apocalypse but disagrees with a man using violence against a woman for the same purpose. Finally, these biases are implicit, as they do not emerge when GPT-4 is directly asked to rank moral violations. These results highlight the necessity of carefully managing inclusivity efforts to prevent unintended discrimination.