{"slug": "ai-coding-agents-generate-more-code-but-not-more-software", "title": "AI coding agents generate more code, but not more software", "summary": "A study of coding practices at more than 700 software firms found that introducing AI coding agents produced a 30 percent increase in lines of code, a 20 percent rise in commits and a 23 percent increase in pull requests, but no statistically significant change in the resolution rate of Jira-tracked Issues and Epics. Harvard University researchers Fiona Chen and James Stratton analyzed 300 million work events from Jellyfish covering over 700,000 employees from 2021 through March 6, 2026, and concluded that human code review absorbs the coding-phase efficiency gains, with longer reviews, more revision-requiring pull requests and more reviewer comments. The researchers reported \"little evidence that firms increase software output or reduce employment\" from AI coding tools.", "body_md": "Anyone who’s even tangentially associated with computer programming knows that modern AI coding assistants and agents [can be incredibly efficient at generating huge amounts of functional code](https://arstechnica.com/ai/2026/01/developers-say-ai-coding-tools-work-and-thats-precisely-what-worries-them/). But coders making use of those tools also [know better than to trust the accuracy of that code](https://arstechnica.com/ai/2025/07/developer-survey-shows-trust-in-ai-coding-tools-is-falling-as-usage-rises/), meaning [substantial effort needs to be spent](https://arstechnica.com/ai/2025/05/time-saved-by-ai-offset-by-new-work-created-study-suggests/) reviewing any AI-generated output.\n\n[A recent study](https://fion.ac/jellyfish.pdf) of actual coding practices across hundreds of firms finds that human code review forms a significant “bottleneck” for the overall efficiency of AI coding tools, resulting in “little evidence that firms increase software output or reduce employment” by using them. Any efficiency increased during the actual coding phase, the study authors find, is “absorbed by downstream constraints in the production process”; as “the code review process significantly increases in length, pull requests are more likely to require revisions, and reviewers leave more comments.”\n\n## Cut once, measure twice\n\nTo come to these conclusions, Harvard University researchers Fiona Chen and James Stratton made use of aggregated analytics data from [Jellyfish](https://jellyfish.co/), which measures the granular output of engineering teams. That data encompasses 300 million individual “work events” (e.g. commits and pull requests) and issue management software data across more than 700,000 employees at over 700 relevant software development firms from 2021 through March of 2026.\n\nTo assess the impact of AI tools on these firms, the researchers used a mix of directly measured AI usage and analyses of Github activity to determine when each company started introducing either AI coding assistants (which can help auto-complete code primarily authored by humans) and/or AI coding agents (which primarily write and submit code autonomously based on prompts) into their workflows. The researchers then perform some complicated math to determine a “difference of differences” regression on key variables both before and after the introduction of these tools at different points in time across different organizations.\n\nIn terms of raw code being produced, the results are clear and stark. The introduction of AI coding agents at a firm leads to a 30 percent increase in total lines of code generated, a 20 percent rise in the number of total commits, and a 23 percent increase in pull requests on average, the researchers write. But all that extra code doesn’t translate directly into improved software output on the firm level. On the contrary, the resolution rate for Issues and Epics (i.e. wholesale software features) tracked by tools like Jira did not change in a statistically significant way after AI tools were introduced (the researchers also found no “compositional shift” in the size or complexity of those Jira-tracked issues across the AI introduction).", "url": "https://wpnews.pro/news/ai-coding-agents-generate-more-code-but-not-more-software", "canonical_source": "https://arstechnica.com/ai/2026/10/ai-coding-agents-generate-more-code-but-not-more-software/", "published_at": "2026-10-09 19:43:50+00:00", "updated_at": "2026-10-09 19:51:43.245219+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-agents", "ai-tools", "developer-tools", "ai-research"], "entities": ["Fiona Chen", "James Stratton", "Harvard University", "Jellyfish", "Jira", "GitHub"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/ai-coding-agents-generate-more-code-but-not-more-software", "markdown": "https://wpnews.pro/news/ai-coding-agents-generate-more-code-but-not-more-software.md", "text": "https://wpnews.pro/news/ai-coding-agents-generate-more-code-but-not-more-software.txt", "jsonld": "https://wpnews.pro/news/ai-coding-agents-generate-more-code-but-not-more-software.jsonld"}}