Large-Scale Empirical Study Analyzes AI-Generated Code in Real-World Repositories

Researchers conduct a large-scale empirical study of code produced by artificial intelligence. The investigation targets code found in real-world software repositories. The study quantifies how often AI-generated code appears in these projects. It examines characteristics and patterns of the AI-produced code. The analysis aims to reveal adoption trends among developers. Findings are intended to inform both practitioners and tool creators. The work contributes to understanding the impact of AI on software development. The paper is publicly available on arXiv.