🔍 Read the full analysis: Why Reviewing Pull Requests Is The New Challenge As AI Agents Write More Code on ThorstenMeyerAI.com
Get hardware and tech essentials 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 report says more than 40% of pull requests on the platform in September 2026 began with a detectable coding agent, up from 17% in March. As AI-generated changes spread, GitHub introduced review and model-routing tools, while keeping approval and accountability rules that require human involvement. The share is a floor, and a merged pull request is not proof that its code is correct.
More than 40% of GitHub pull requests in September 2026 started with a detectable coding agent, up from 17% in March, according to the platform’s Octoverse report released October 28. The growth is shifting attention from how quickly AI can draft code to whether developers and their organizations can review, approve and take responsibility for it.
GitHub counted 110 million agent-started pull requests in September, roughly six times the March volume. It defines an agent-started pull request as one whose first draft came from a coding agent GitHub can detect. The company describes the count as a floor because not all agent use is identifiable. The reported share includes public and private repositories; GitHub did not provide a split between them.
In September, 90% of agent pull requests were merged within two weeks, compared with 83% in February. Pull requests without a detected agent held at about 86%. Those figures track whether a change was merged, not whether it was correct, secure or free of later rework. GitHub also reported that 87% of agent pull requests changed more than one file, compared with 79% of pull requests without a detected agent.
GitHub’s announcements focused on review and routing. Copilot code review Max brings together reviews from several AI model providers into one combined review; the company also made Copilot approvals count toward required approvals and released Copilot Fusion, a tool that routes tasks to models. GitHub estimates an average review costs $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
Review Capacity Becomes the Constraint
The figures point to a change in where software teams may spend time. When agents produce more proposed changes, the work of understanding a diff, checking assumptions and deciding whether it is safe to merge can become a constraint. That is particularly relevant when a pull request touches multiple files or requires knowledge of how separate parts of a system interact.
There is also a cost and governance question. Review credits and Actions usage add to the expense of agent-assisted development, while teams still need enough skilled reviewers to catch mistakes. Companies may need to track rework, reversions and incidents, not just the volume of pull requests or how quickly they merge. For developers, task definition and critical review matter alongside producing code.
The trend may affect early-career work as well: small, bounded coding tasks have often helped junior developers learn a codebase. The source report raises this concern, but does not quantify how agent use is changing hiring, training or junior responsibilities. The impact on those pathways remains an open question.
Products Worth Considering
LG 24G414B-B 24-Inch UltraGear Full HD IPS Gaming Monitor
LG UltraGear 27G640A-B 27-Inch QHD IPS Gaming Monitor, 300Hz, 1ms, G-SYNC Compatible, FreeSync Premium, DisplayHDR 400
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
As an affiliate, we earn on qualifying purchases.
How GitHub Defines Agent Work
GitHub’s 2026 Octoverse edition covers October 1, 2025, through September 30, 2026 and reports activity on its own platform. The agent measure is based on the first draft and a signal GitHub can detect; a human may subsequently revise much of the code and the pull request still counts. GitHub’s report uses several related labels, including “agent-started” and “agent-authored,” but “agent-started” most closely matches the stated definition.
The report says GitHub recalculated 2024 and 2025 figures using updated definitions and a revised window, so comparisons with earlier Octoverse reports may not be direct. Platform activity also grew broadly: GitHub reported 1.3 billion merged pull requests during the 12-month reporting period, compared with 538 million in 2025 and 402 million in 2024. Those totals describe all pull requests, not only agent-started ones.
GitHub’s broader figures show increased use of its agent features: 55% of monthly active Copilot users used agent features, compared with 35% a year earlier, while use of its cloud agent grew 230% year over year. These are company-reported platform metrics, not an independent measure of code quality or productivity across the software industry.
“A pull request whose first draft came from a coding agent that we can detect.”
— GitHub, in its Octoverse report
What the Metrics Do Not Show
The reported 40%-plus share is not a complete count: GitHub says it detects only some agent use, and the actual share could be higher. The company has not published a public-versus-private repository breakdown for this measure. Nor does its reported merge rate establish that the resulting code works correctly or avoids defects.
The available figures do not show how much review time agent changes require, how often reviewers catch serious problems, or whether AI-assisted reviews reduce errors. GitHub’s cost estimates are averages for its review tiers, not a full estimate of team spending, and the supplied report does not detail how the stated figures vary by workload. The long-term effect on developer roles and training is also not quantified.
Teams Face Review Decisions
GitHub has released its review and routing tools, but organizations will decide whether and where to use AI approvals in their own workflows. They will also need to determine how many reviews to run, who can approve a change, and how to budget for review credits and Actions minutes.
The next useful measure will be whether higher agent pull-request volume leads to more rework, reversions or production incidents, alongside any changes in review speed. GitHub’s report establishes the scale of detected agent activity on its platform; it does not settle whether the new tools can keep review quality aligned with that growth.
Key Questions
What does GitHub mean by an agent-started pull request?
GitHub defines it as a pull request whose first draft came from a coding agent it can detect. A person can edit the draft afterward and it still counts under that definition.
Does the 40%-plus figure cover every AI-generated pull request?
No. GitHub calls the metric a floor because its systems may not detect all agent use. The company did not publish a separate figure for public and private repositories.
Does a high merge rate show that agent code is correct?
No. The reported 90% figure measures the share of agent pull requests merged within two weeks. It does not establish code quality, correctness or whether changes caused later problems.
How much does Copilot code review cost?
GitHub estimates an average review at $0.25 for Lite, $1.50 for Balanced and $10 for Max, plus GitHub Actions minutes. These are GitHub’s stated estimates, not a full accounting of all costs a team may incur.
Can GitHub’s cloud agent approve its own pull request?
No. GitHub says its cloud agent cannot approve or merge its own pull requests, and the person who requested a change cannot approve it either.
Source: ThorstenMeyerAI.com
