# A Framework for Identifying, Categorizing, and Explaining Bias in AI-Generated Code

arXiv: https://arxiv.org/abs/2609.30642 · HTML: https://mlsanity.com/paper/2609.30642/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 45 references that are in the map.
- First posted on arXiv 25 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [Equality of Opportunity in Supervised Learning](https://mlsanity.com/paper/1610.02413/) (2016)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [A Reductions Approach to Fair Classification](https://mlsanity.com/paper/1803.02453/) (2018)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP](https://mlsanity.com/paper/2103.00453/) (2021)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI](https://mlsanity.com/paper/2201.08164/) (2022)
- [What's in a Name? Reducing Bias in Bios without Access to Protected Attributes](https://mlsanity.com/paper/1904.05233/) (2019)
- [Representation Bias in Data: A Survey on Identification and Resolution Techniques](https://mlsanity.com/paper/2203.11852/) (2022)
- [Investigating Reasons for Disagreement in Natural Language Inference](https://mlsanity.com/paper/2209.03392/) (2022)
- [Dbias: Detecting biases and ensuring Fairness in news articles](https://mlsanity.com/paper/2208.05777/) (2022)
- [Uncovering and Quantifying Social Biases in Code Generation](https://mlsanity.com/paper/2305.15377/) (2023)
- [Bias Unveiled: Investigating Social Bias in LLM-Generated Code](https://mlsanity.com/paper/2411.10351/) (2024)
- [How Developers Interact with AI: A Taxonomy of Human-AI Collaboration in Software Engineering](https://mlsanity.com/paper/2501.08774/) (2025)
- [The Invisible Hand: Unveiling Provider Bias in Large Language Models for Code Generation](https://mlsanity.com/paper/2501.07849/) (2025)
- [Towards Multi-dimensional Explanation Alignment for Medical Classification](https://mlsanity.com/paper/2410.21494/) (2024)
- [Bias-Aware Agent: Enhancing Fairness in AI-Driven Knowledge Retrieval](https://mlsanity.com/paper/2503.21237/) (2025)
- [In-Context Learning (and Unlearning) of Length Biases](https://mlsanity.com/paper/2502.06653/) (2025)
- [Are Prompts All You Need? Evaluating Prompt-Based Large Language Models (LLM)s for Software Requirements Classification](https://mlsanity.com/paper/2509.13868/) (2025)
- [Human-AI Collaboration and the Transformation of Software Engineering Work](https://mlsanity.com/paper/2606.03394/) (2026)
- [Towards clinical AI fairness: A translational perspective](https://mlsanity.com/paper/2304.13493/) (2023)

## Abstract (arXiv)

As Large Language Models (LLMs) become integrated into software development workflows, concerns regarding unintentional biases in AI-generated code. Although evidence suggests these biases exist, limited research has systematically identified, categorized, and explained them. This study investigates bias in AI-generated code and evaluates whether LLMs can reliably identify and explain it through a taxonomy-driven framework. We extended an existing dataset of biased AI-generated Python code and manually annotated snippets with bias categories and human-authored justifications to establish a ground-truth dataset. Using this dataset, we evaluated proprietary and open-source LLMs as automated bias detection and justification systems through ICL. Finally, we analyzed similarity between LLM-generated explanations and human-authored justifications using structured justification and code identification metrics. Our findings demonstrate that LLMs can effectively support code bias identification and explanation. Gemini achieved 80.14% classification accuracy, with 84.0% precision and 95.7% recall, while the best open-source alternative, Qwen3-coder, achieved 82.45% accuracy, 68.64% precision, and 80.22% recall. Additionally, the models achieved justification similarity scores of 80.4% and 80.14%, respectively, relative to human-authored reasoning, and code identification similarity scores of 86.0% and 87.82%. These results suggest that LLMs can detect biased logic in generated Python code and produce explanations that substantially align with expert interpretations.