Study Finds 46% of AI-Generated Code Fixes Are Rejected by Developers
A new study analyzing the AIDev dataset found that 46.41% of pull requests generated by AI coding agents — including Copilot, Devin, Cursor, and Claude — are rejected by human developers. Researchers examined 306 non-merged pull requests and identified 14 distinct reasons for rejection, grouped into four categories: incorrect implementation, CI/test failures, agent inability to complete the task, and low priority. The findings highlight significant inefficiencies in current AI-assisted software development workflows and point to concrete areas for improvement.
Researchers from a study accepted to the 2026 Mining Software Repositories (MSR) conference conducted a qualitative and quantitative analysis of rejected AI-generated pull requests using the AIDev dataset. Across four prominent AI coding agents — GitHub Copilot, Devin, Cursor, and Claude — nearly half of all proposed code fixes were ultimately discarded without being merged. The 14 identified rejection reasons fall into four high-level categories: implementation incorrectness (e.g., wrong approach or incomplete code), failure to pass continuous integration pipelines and automated tests, agent inability to produce any working output (e.g., lost sessions or no code generated), and low prioritization of the underlying issue. The authors argue that these rejections represent a substantial waste of resources, including human review time, compute cycles, and API token usage. To address these failure modes, the paper recommends better upfront guidance to agents about acceptable approaches, explicit constraints on disallowed strategies, and clearer instructions for validating implementations against CI pipelines. The study frames improved agent-developer collaboration as essential for making AI coding tools genuinely efficient teammates rather than sources of additional review burden.
What's missing
The study does not report whether rejection rates differ significantly across the four individual agents (Copilot, Devin, Cursor, Claude), which would be important for understanding whether some tools perform substantially better than others. Additionally, the dataset's time range and how representative it is of current agent capabilities is not specified, and the study does not account for whether some rejections reflect project-specific norms rather than agent failure.
What different sources said
- arXiv cs.AICenter
Understanding the Rejection of Fixes Generated by Agentic Pull Requests -- Insights from the AIDev Dataset
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.