🔍 Read the full analysis: The Review Side Of GitHub Pull Requests Is Changing With AI Agents on ThorstenMeyerAI.com
Get privacy and security gear 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 its platform in September 2026 began with a detectable coding agent, up from 17% in March. The company introduced review and routing features as agent-written changes increase, while saying developers remain accountable for code that ships. GitHub’s measure is a floor, and a merged pull request is not proof that 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 GitHub’s Octoverse report released October 28. Alongside the figures, the company introduced features for AI-assisted code review and model routing, addressing a shift in which producing a proposed change can be faster than checking it before it is merged.
GitHub says the September figure represents 110 million agent-started pull requests in a month, about six times the March volume. The company defines an agent-started pull request as one whose first draft came from a coding agent it can detect. It calls the figure a floor because some agent use may not be detected. A pull request can still count if a person substantially edits the draft afterward.
GitHub also reported that 90% of detected agent pull requests were merged within two weeks in September, compared with 83% in February. The rate for pull requests without a detected agent was about 86%. These figures describe whether a change was merged, not whether it was correct, safe or free of later rework.
The new offerings focus on review. Copilot code review Max gathers feedback from several AI reviewers using different model providers and returns a combined review. GitHub also made Copilot approvals eligible to count toward required approvals and released Copilot Fusion, a system that routes work to models for different steps. The reported average price per review is $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 Becomes a Paid Workflow
The figures point to a practical change for development teams: as agents generate more proposed changes, review capacity and attention become a larger part of the work. AI review may help surface issues or organize feedback, but teams still need people to judge whether the change fits the system, meets requirements and is safe to merge.
The costs also make review choices a budgeting decision. GitHub’s estimates vary substantially by review tier, and Actions minutes add another charge. Companies will need to decide which repositories or changes warrant more intensive review, whether AI-generated feedback can count toward formal approval, and how to track outcomes such as rework, reversions and incidents rather than treating pull-request volume as a measure of quality.
For developers, the shift may change which tasks build experience. Small, well-defined coding jobs have often offered a way to learn a codebase. If agents take on more first drafts, developers may spend more time specifying tasks and scrutinizing diffs. That is a plausible workforce implication, not a finding established by the reported pull-request statistics.
Products Worth Considering
Logitech MX Mechanical Wireless Illuminated Keyboard Tactile (Graphite)
LG UltraGear 32GX850A-B 32-Inch 4K UHD Glossy OLED Gaming Monitor
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
The Octoverse 2026 report covers October 1, 2025, through September 30, 2026. GitHub says its platform metrics generally include public and private repositories unless a metric is explicitly excluded. The agent-share figure has no stated exclusion, but GitHub did not provide a public-versus-private breakdown in the supplied report material.
GitHub’s wording varies across the report: it describes pull requests as “agent-started,” “agent-generated” or “agent-authored.” Its definition focuses on the first draft and detectable signals, rather than measuring how much of the final code an agent wrote. The company recalculated its 2024 and 2025 figures using updated definitions and a revised reporting window, so those historical numbers are not directly comparable with earlier Octoverse reports.
The wider platform also saw substantial activity. GitHub reported 1.3 billion merged pull requests during the reporting year, compared with 538 million in 2025 and 402 million in 2024. It said 55% of monthly active Copilot users used agent features, up from 35% a year earlier. Those platform totals provide scale, but do not by themselves establish that agents caused every increase.
“A pull request whose first draft came from a coding agent that we can detect.”
— GitHub, in its Octoverse report
What the Metrics Leave Out
The report does not give a public-versus-private repository split for agent-started pull requests, nor does it quantify how much of each final change was written or altered by a person. Because detection is incomplete, the reported share should not be read as a complete count of all agent use. The wording differences across the report also make “started with an agent’s first draft” the clearest description when citing the figure.
The reported merge rates do not show whether changes later caused defects, needed extensive rework or were reverted. The supplied material also does not establish whether the new review features improve code quality or reduce review time. GitHub says its cloud agent cannot approve or merge its own pull requests, and that the person who requested a change cannot approve it; details about how organizations configure AI approvals and apply those rules across workflows remain limited here.
Teams Weigh Review Rules and Cost
As the new features roll out, teams will have to set policies for when AI review is appropriate and whether an AI approval can satisfy a required review step. They will also need to monitor usage charges, including the review tier and Actions minutes, alongside human reviewer time.
The next useful test will be how these workflows perform in practice: whether they catch actionable problems, change review turnaround, or affect rework and production incidents. GitHub’s October report establishes the growth in detectable agent-started pull requests and outlines its new tools, but it does not yet answer those outcome questions. Future platform data and company-level results will help show whether faster drafting can be matched by dependable review.
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 that the company can detect. Human edits afterward do not necessarily remove it from the count.
How many pull requests did agents start in September?
GitHub reports 110 million in September 2026, representing more than 40% of its pull requests that month. It describes the figure as a floor because not all agent use may be detected.
Does a high merge rate mean agent-written code is correct?
No. GitHub reported that 90% of detected agent pull requests were merged within two weeks, but merging is not a measure of correctness or of whether a change later requires fixes.
How much does GitHub’s AI code review cost?
GitHub estimates an average review at $0.25 for Lite, $1.50 for Balanced and $10 for Max, with GitHub Actions minutes charged separately.
Who remains responsible for code that ships?
GitHub says developers remain accountable. 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
