🔍 Read the full analysis: Is Pull Request Review Becoming The Main Challenge For AI-Assisted Teams? on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
GitHub’s Octoverse 2026 report says more than 40% of pull requests on its platform in September began with a coding agent’s first draft, up from 17% in March. The figures point to review and approval as growing concerns, but they do not show whether merged agent-started code is correct or how the trend varies between public and private repositories.
More than 40% of pull requests on GitHub in September began with a coding agent’s first draft, up from 17% in March, according to the company’s 2026 Octoverse report, released October 28. The growth puts more proposed code into review and comes as GitHub launches tools for AI-assisted review and approval, but the reported merge rates do not establish that the code was correct or safe.
GitHub says the September share represents 110 million agent-started pull requests in one month, about six times the volume it recorded in March. The company defines these as pull requests whose first draft came from a coding agent it can detect. It calls the count a floor because some agent use may go undetected. The report includes public and private repositories unless otherwise specified, but does not break this figure out by repository type.
Most detected agent-started pull requests were merged: GitHub reports that 90% were merged within two weeks in September, compared with 83% in February. Pull requests without a detected agent had a merge rate of about 86%, according to the report. A merge records that a change was accepted into a codebase; it does not, by itself, measure defects, later rework or effects in production.
At its Universe conference, GitHub announced additions aimed at review and routing. Copilot code review Max combines reviews from multiple model providers into one result, while Copilot Fusion routes tasks to models. GitHub also said Copilot approvals can count toward required approvals under supported workflows. The company estimates an average review will cost $0.25 for Lite, $1.50 for Balanced and $10 for Max, plus GitHub Actions minutes.
Infographic 01 · The agent share
Four in ten pull requests now start with an agent.
GitHub’s own telemetry, March to September 2026. The first draft comes from an agent; review and merge stay with people.
Merged within two weeks
Share of pull requests merged within 14 days
Merged fast is not the same as correct. GitHub publishes no revert or defect rate.
Bigger changes to review
Share of pull requests that change more than one file
More files per change means more for a reviewer to check.
What GitHub counts: an “agent-started pull request” is one “whose first draft came from a coding agent that we can detect.” Undetected agents are missing, so GitHub calls the figure a floor. Metrics cover public and private repositories unless noted.
Source: GitHub Octoverse 2026, 28 Oct 2026 (text, chart captions, glossary and methodology). All figures are GitHub’s own.
Powered by Thorsten Meyer AI
More Code Means More Review Work
The reported increase changes the workload facing teams: when agents produce more proposed changes, people still need to judge whether those changes fit the task, work with surrounding systems and are safe to merge. GitHub’s report says 87% of agent-started pull requests changed more than one file, compared with 79% of pull requests without a detected agent. That difference may mean reviewers have more connected changes to understand, though the statistic does not measure time spent reviewing or defect rates.
The new review products also give teams a price to weigh against that workload. GitHub’s estimates range from 25 cents to $10 per review, before Actions usage. Actual spending will depend on how often teams invoke reviews and which tier they choose. For organizations, the practical issue is not simply how many pull requests agents produce, but whether review capacity, budgets and checks keep pace.
GitHub says developers remain accountable for code that ships. Its cloud agent cannot approve or merge its own pull request, and the person who requested a change cannot approve it either. That division leaves human approval in the process, even as AI-generated reviews may count toward required approvals in some configurations. Teams will need to decide where that automated input is appropriate and how to measure results beyond merge totals.
Products Worth Considering
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
LG 34GX90SA-W 34-inch Ultragear WQHD OLED Curved Gaming Monitor
Samsung 27 Odyssey OLED G5 (G50SF) QHD QD-OLED Gaming Monitor
As an affiliate, we earn on qualifying purchases.
How GitHub Defines Agent-Started
The Octoverse report is GitHub’s annual account of activity on its platform. The 2026 edition covers October 1, 2025, through September 30, 2026. Its agent measure depends on detection: a pull request counts if GitHub can identify a coding agent as the source of the first draft. A human may subsequently change that draft, and the pull request still counts as agent-started.
GitHub’s report uses several related labels, including “agent-started,” “agent-generated” and “agent-authored.” For the headline statistic, “agent-started” most closely reflects the company’s definition. GitHub also recalculated its 2024 and 2025 figures using the new definitions and measurement window, so comparisons with earlier Octoverse reports may not be direct.
The wider platform saw substantial growth during the same period. GitHub reports 1.3 billion merged pull requests in the year to September, compared with 538 million in 2025 and 402 million in 2024. Those totals describe overall platform activity; they do not establish that agents alone caused the increase. GitHub also says 55% of Copilot’s monthly active users used agent features, up from 35% a year earlier.
““a pull request whose first draft came from a coding agent that we can detect.””
— GitHub, in its Octoverse 2026 report
What the Merge Figures Cannot Show
The report does not establish whether the higher merge rate for agent-started changes reflects code quality, team review practices or other factors. Merged is not the same as correct, and GitHub’s reported rates do not provide a direct comparison of bugs, security incidents, reversions or follow-up work. The source material does not give those outcome measures for the agent-started pull requests.
Several details also remain unavailable: GitHub has not provided a public-versus-private split for the agent share, and the company’s detection limits mean the count is incomplete by its own account. The report’s wording varies across sections, so the exact claim is that the first draft came from a detectable coding agent—not necessarily that the agent wrote every line ultimately merged.
GitHub’s prices are estimates rather than a guaranteed bill for every review. The final cost depends on usage and associated Actions minutes. The available figures also do not show how many teams will adopt the new review features or whether AI-generated approvals will change review quality or speed.
How Teams Will Measure Review Quality
GitHub’s report sets out the September figures and its new review offerings, but the next useful evidence will be about outcomes: how often agent-assisted changes require rework, are reverted or lead to incidents, and how much human review time they take. Companies adopting these tools can track those measures alongside review costs and merge rates, rather than treating more pull requests as a measure of success.
GitHub has not supplied a date in the source material for a follow-up outcome study or for wider changes to the review features. For now, teams must set their own approval rules, choose where AI review can count and keep human responsibility for code that ships. Further platform reporting may show whether the rapid rise in agent-started changes is matched by better review capacity and dependable results.
Key Questions
What does GitHub mean by an agent-started pull request?
GitHub counts a pull request when its first draft came from a coding agent the company can detect. Later human edits do not change that classification.
How many pull requests did agents start in September?
GitHub reports 110 million agent-started pull requests in September 2026, representing more than 40% of pull requests on its platform that month.
Does a high merge rate prove agent-written code is reliable?
No. GitHub reports that 90% of detected agent-started pull requests were merged within two weeks, but a merge does not establish correctness. The report does not provide comparative defect or incident rates for these changes.
How much does GitHub estimate an AI review costs?
GitHub estimates an average Copilot review at $0.25 for Lite, $1.50 for Balanced and $10 for Max, with GitHub Actions minutes charged separately.
Who is responsible for code an agent helps produce?
GitHub says developers remain accountable for code that ships. Its cloud agent cannot approve or merge its own pull request, and the person who requested the change cannot approve it either.
Source: ThorstenMeyerAI.com
