Search papers, labs, and topics across Lattice.
This paper investigates the integration of AI-assisted pull requests (PRs) in open-source software (OSS) by expanding the AIDev dataset with contributor code ownership data and a human-created PR baseline. The study reveals that a majority of AI-co-authored PRs are submitted by contributors without prior code ownership, while most repositories lack guidelines for AI coding agent usage. The key finding is that AI-co-authored PRs are merged faster with significantly less feedback, especially those from non-owner developers, suggesting a potential lack of thorough review.
AI-generated code in open source gets a free pass: 80% of AI-co-authored pull requests from new contributors are merged without review, raising questions about code quality and security.
Large Language Models (LLMs) increasingly automate software engineering tasks. While recent studies highlight the accelerated adoption of ``AI as a teammate''in Open Source Software (OSS), developer interaction patterns remain under-explored. In this work, we investigated project-level guidelines and developers'interactions with AI-assisted pull requests (PRs) by expanding the AIDev dataset to include finer-grained contributor code ownership and a comparative baseline of human-created PRs. We found that over 67.5\% of AI-co-authored PRs originate from contributors without prior code ownership. Despite this, the majority of repositories lack guidelines for AI-coding agent usage. Notably, we observed a distinct interaction pattern: AI-co-authored PRs are merged significantly faster with minimal feedback. In contrast to human-created PRs where non-owner developers receive the most feedback, AI-co-authored PRs from non-owners receive the least, with approximately 80\% merged without any explicit review. Finally, we discuss implications for developers and researchers.
