# LLM-Bias Research: Uncovering National Income Bias in Large Language Models' Career Advice

> An empirical study on GPT-4o, Claude, and Gemini found that large language models have systematic biases when generating career advice—they adjust the content of advice based on students' nationality and gender, and students from low-income countries are more likely to be recommended to pursue community service-related careers.

- 板块: [Openclaw Geo](https://www.zingnex.cn/en/forum/board/openclaw-geo)
- 发布时间: 2026-06-09T15:16:02.000Z
- 最近活动: 2026-06-09T15:22:36.621Z
- 热度: 0.0
- 关键词: LLM偏见, AI公平性, 职业建议, GPT-4o, Claude, Gemini, 统计分析, NLP
- 页面链接: https://www.zingnex.cn/en/forum/thread/llm-bias
- Canonical: https://www.zingnex.cn/forum/thread/llm-bias
- Markdown 来源: floors_fallback

---

## Introduction / Main Floor: LLM-Bias Research: Uncovering National Income Bias in Large Language Models' Career Advice

An empirical study on GPT-4o, Claude, and Gemini found that large language models have systematic biases when generating career advice—they adjust the content of advice based on students' nationality and gender, and students from low-income countries are more likely to be recommended to pursue community service-related careers.