John Businge

Daniel Ogenrwot, John Businge

AI coding agents are increasingly acting as autonomous contributors by generating and submitting pull requests (PRs). However, we lack empirical evidence on how these agent-generated PRs differ from human contributions, particularly in how they modify code and describe their changes. Understanding these differences is essential for assessing their reliability and impact on development workflows. Using the MSR 2026 Mining Challenge version of the AIDev dataset, we analyze 24,014 merged Agentic PRs (440,295 commits) and 5,081 merged Human PRs (23,242 commits). We examine additions, deletions, commits, and files touched, and evaluate the consistency between PR descriptions and their diffs using lexical and semantic similarity. Agentic PRs differ substantially from Human PRs in commit count (Cliff's $δ= 0.5429$) and show moderate differences in files touched and deleted lines. They also exhibit slightly higher description-to-diff similarity across all measures. These findings provide a large-scale empirical characterization of how AI coding agents contribute to open source development.

Daniel Ogenrwot, John Businge

Software Engineering 3.0 marks a paradigm shift in software development, in which AI coding agents are no longer just assistive tools but active contributors. While prior empirical studies have examined productivity gains and acceptance patterns in AI-assisted development, the challenges associated with integrating agent-generated contributions remain less understood. In particular, merge conflicts, a fundamental aspect of collaborative software development, remain underexplored in this context. In this paper, we present AgenticFlict, a large-scale dataset of textual merge conflicts in AI coding agent pull requests (Agentic PRs). The dataset comprises 142K+ Agentic PRs collected from 59K+ repositories, of which 107K+ are successfully processed through deterministic merge simulation. Our pipeline identifies 29K+ PRs exhibiting merge conflicts, yielding a conflict rate of 27.67%, and extracts 336K+ fine-grained conflict regions across these instances. Our preliminary exploratory analysis indicates that merge conflicts are both frequent and often substantial in AI-generated contributions, with noticeable variation across agents, emphasizing the need to better understand and manage integration challenges in AI-assisted software development. The dataset, code and supplementary materials are available in zenodo: https://doi.org/10.5281/zenodo.19396917.

Jorge Gonzalo Delgado Cervantes, John Businge, Daniel Ogenrwot

Clone-and-own development produces families of related software variants that evolve independently. As variants diverge, important fixes applied in one repository are often missing in others. PaReco has shown that thousands of such missed opportunity (MO) patches exist across real ecosystems, yet its textual output provides limited support for understanding where and how these fixes should be propagated. We present MOVis, a lightweight, interactive desktop tool that visualizes MO patches between a source and target variant. MOVis loads PaReco's MO classifications and presents patched and buggy hunks side-by-side, highlighting corresponding regions and exposing structural differences that hinder reuse. This design enables developers to quickly locate missed fixes, understand required adaptations, and more efficiently maintain consistency across software variants. The tool, replication package, and demonstration video are available at https://zenodo.org/records/18356553 and https://youtu.be/Ac-gjBxHJ3Y.

Daniel Ogenrwot, John Businge

The rapid adoption of large language models (LLMs) like ChatGPT has introduced new dynamics in software development, particularly within pull request workflows. While prior research has examined the quality of AI-generated code, less is known about how developers evaluate, adapt, and integrate these suggestions in real-world collaboration. We analyze 338 pull requests from 255 GitHub repositories containing self-admitted ChatGPT usage, comprising 645 AI-generated snippets and 3,486 developer-authored patches. To support this analysis at scale, we use PatchTrack, an automated classifier that identifies whether AI-generated patches were applied, partially reused, or not integrated. Our findings reveal that full adoption of ChatGPT-generated code is uncommon: the median integration rate is 25%. Qualitative analysis of 89 pull requests with integrated patches reveals recurring patterns of structural integration, selective extraction, and iterative refinement, indicating that developers typically treat AI output as a starting point rather than a final implementation. Even when code is not directly adopted, ChatGPT influences workflows through conceptual guidance, documentation, and debugging strategies. Integration decisions reflect contextual fit, integration effort, maintainer trust, and established pull request review norms rather than serving as direct indicators of code correctness. Overall, this study provides empirical insight into AI-mediated decision-making in collaborative software development, showing that the influence of generative AI extends beyond patch generation to how developers reason about, adapt, and negotiate code during review within pull request workflows. These findings inform the design of AI-assisted tools and support more transparent and effective use of LLMs in practice.