Date
Publisher
arXiv
The popularization of AI chatbot usage globally has created opportunities for research into their benefits and drawbacks, especially for students using AI assistants for coursework support. This paper asks: how do LLMs perceive the intellectual capabilities of student profiles from intersecting marginalized identities across different cultural contexts? We conduct one of the first large-scale intersectional analyses on LLM explanation quality for Indian and American undergraduate profiles preparing for engineering entrance examinations. By constructing profiles combining multiple demographic dimensions including caste, medium of instruction, and school boards in India, and race, HBCU attendance, and school type in America, alongside universal factors like income and college tier, we examine how quality varies across these factors. We observe biases providing lower-quality outputs to profiles with marginalized backgrounds in both contexts. LLMs such as Qwen2.5-32B-Instruct and GPT-4o demonstrate granular understandings of context-specific discrimination, systematically providing simpler explanations to Hindi/Regional-medium students in India and HBCU profiles in America, treating these as proxies for lower capability. Even when marginalized profiles attain social mobility by getting accepted into elite institutions, they still receive more simplistic explanations, showing how demographic information is inextricably linked to LLM biases. Different models (Qwen2.5-32B-Instruct, GPT-4o, GPT-4o-mini, GPT-OSS 20B) embed similar biases against historically marginalized populations in both contexts, preventing profiles from switching between AI assistants for better results. Our findings have strong implications for AI incorporation into global engineering education.
Study at a glance
- Question: Do large language models alter STEM explanation complexity based on students' intersecting demographic identities across different national contexts?
- Setting and data: Simulated undergraduate STEM education contexts across the United States and India, evaluating 200 synthetic student profiles on 100 math and science benchmark problems.
- Method: Algorithmic audit using synthetic persona experiments across ranking and generation tasks, evaluated with readability indices and SHAP feature attribution.
Why it matters
Personalized educational AI may judge student capability using demographic attributes rather than demonstrated ability. Overlooking Global South contexts and intersecting identities leaves these potential instructional disparities untested at scale.
Key findings
- Language models systematically provided simpler STEM explanations to marginalized student profiles, creating a gap of up to 2.55 grade levels between the most privileged and most marginalized identities.
- Lower student income, non-English medium of instruction in India, and disability status consistently prompted models to generate lower-complexity instructional content across both countries.
- Disparities compounded across multiple intersecting identities and persisted within elite institutions, where low-income or caste-oppressed students at premier engineering colleges still received simpler explanations.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
What is the application?
Who is the user?
Who age?
Why use AI?
Study design
