To What Extent Does Agent-Generated Code Require Maintenance? An Empirical Study

An empirical study posted to arXiv on 7 May 2026 by Tatsuya Shirai found that AI-generated code files receive less frequent maintenance than human-authored code, with updates touching only a small fraction of file size. Analyzing over 1,000 files and roughly 3,200 changes across 100 popular repositories using the AIDev dataset of AI-generated pull requests and GitHub, the study found the most frequent modifications to AI code are feature extensions while human updates focus on bug fixes, and that human developers perform the large majority of this maintenance.

Computer Science Software Engineering Submitted on 7 May 2026 v1 https://arxiv.org/abs/2605.06464v1 , last revised 9 May 2026 this version, v2 Title:To What Extent Does Agent-generated Code Require Maintenance? An Empirical Study View PDF https://arxiv.org/pdf/2605.06464 HTML experimental https://arxiv.org/html/2605.06464v2 Abstract:LLM-based autonomous coding agents have reshaped software development. While these agents excel at code generation, open questions persist about the long-term maintainability of AI-generated code. This study empirically investigates the maintenance extent, human involvement, and modification types of AI-generated files versus human-authored code. Using the AIDev dataset of AI-generated pull requests and GitHub, we analyzed over 1,000 files and approximately 3,200 changes from 100 popular repositories. Our findings show that: i AI-generated files receive less frequent maintenance than human-authored code, with updates affecting only a small fraction of file size; ii the most frequent modifications to AI code are feature extensions, whereas human updates focus on bug fixes, and iii human developers perform the large majority of this maintenance. Submission history From: Tatsuya Shirai view email https://arxiv.org/show-email/b89856e8/2605.06464 Thu, 7 May 2026 15:52:41 UTC 766 KB \ v1\ https://arxiv.org/abs/2605.06464v1 v2 Sat, 9 May 2026 06:52:52 UTC 766 KB Additional Features References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .