Code review has become the new bottleneck
„AI generates more code than humans can review, so review must also be automated with AI-review, risk levels, and automatic approval.”
Counter-narrative
The reliability of AI review is questionable. According to the GhostCommit method, AI-based code review fails to detect certain hidden modifications. A leader at Greptile stated that in-house built AI code reviewers often miss the most critical errors. JetBrains Research proposes a three-level workflow to address trust issues in review, and one headline reports a 46% trust gap. Bjarne Stroustrup warns of AI code errors and the importance of review.
Carriers
GitHub (Copilot code review), Anthropic (Claude Code Review), CodeRabbit, Qodo, Blacksmith, Sourcegraph, Gergely Orosz, The New Stack, investors (review-unicorns)
Analysis
The narrative stems from code generation becoming cheap, shifting focus to expensive verification. Capital follows this shift. CodeRabbit raised $143 million at a $1.5 billion valuation. AI code tester Blacksmith's valuation grew to $1 billion in two months. An AI code verification company became a unicorn with a 10.8 billion yuan valuation. Toolmakers have an interest in making the solution an additional AI layer. This piles machine review on machine code and can blur responsibility. The weak point: the stories examined lack independent measurement of whether fewer bugs reach production with AI review. Meanwhile, blind spots are known (GhostCommit), and a Greptile executive says in-house solutions cost $3–25 per pull request. The GitHub ReviewBench benchmark is a manufacturer research preview. One analysis suggests AI-generated code review burns out senior engineers.
What's at stake: Whether the software quality gate remains human-led or AI audits AI, leaving accountability unassigned.
Narrative sub-threads
The gain is lost in review
Excess code generated by agents does not proportionally accelerate delivery because code review, CI, and validation cannot keep pace, so the review process must adapt.
It is based on research, surveys, and engineering cases that measure the problem itself rather than offering a tool or product.
▲ at peak · evidence: strongPlatform-level AI review and automatic approval
Major developer platforms offer built-in AI code review, making some features generally available, and GitHub Copilot code review can now approve pull requests without human review; this feature is disabled by default.
The actors here are platform manufacturers, and the mechanism is product-embedded, sometimes automatic approval and the spread of bot-to-bot reviews.
↗ rising · evidence: moderateHow reliable is AI review?
AI review finds real bugs and also delayed the release of PostgreSQL 19. However, it has blind spots, so benchmarks and studies are emerging to measure it.
The evidence here is a measurement, attack, and open-source case examining the performance and exploitability of review tools, not the launch or demand.
→ lasting · evidence: moderateInvestors bet on the control layer
Valuations of startups specializing in code review and testing are rising fast, as investors bet on governing and validating AI-generated code.
The news is about financing and evaluation, and the actors are independent review and testing companies, not platform makers or researchers.
Sub-thread stories
The gain is lost in review
Agentic AI rewrites code review economics as Rootly drops its small-PR ruletech leadRules on PR size and review process need rethinking for the volume of change agents produce.senior developerSplit large AI PRs into smaller, stacked changes before review.QA testerReviews can rely on automated evidence (tests, checks) instead of line-by-line reading.Articles urge risk-tiered rules as AI-generated code review burden growstech leadIntroduce a risk-tiered review rule so that large AI-generated diffs do not slip through without being reviewed.senior developerReduce the review load with smaller units of change and automated checks.QA testerTests and acceptance criteria will play a more prominent role in verifying AI-generated code.AI speeds up software development, but the bottleneck shifts to review and operationsCTOProductivity should be measured against finished, live software, not the amount of code generated.QA testerThe larger volume of changes puts more testing load on QA.AI broke code review: experts disagree on what should replace ittech leadThe review process needs redesigning: what humans check, what machines check, and the upper limit on PR size.senior developerReview becomes a core competence, as checking for errors is harder than writing the code.QA testerAutomated test coverage becomes the main line of defense against AI-generated changes.Anthropic describes how agentic coding strains CI and how it scaled test impact analysisQA testerRunning only the affected tests, rather than the full suite, can be the solution.tech leadCI costs driven by agents can be planned for if tests are selected based on changes.After AI code generation, verification is the bottleneck: more commits, not proportionally more releasestech leadInvest in review capacity and automated verification, because writing speed is no longer the constraint.QA testerValidation gains in value, and testing and checking AI code become a central task.CTOMeasure impact by release frequency and defects, not by commit count.200-person study: AI coding speeds up implementation but leaves harder engineering decisionssenior developerPlan more time for review and architectural decisions.tech leadDo not raise expectations automatically in proportion to faster implementation.AI-generated code is burning out senior engineers in code review, analysis sayssenior developerRecurring review feedback should be recorded as rules so that automation can check it.tech leadPlan how review load is shared and how AI reviewers are introduced before reviewers burn out.Harvard study: AI agents write 30% more code but do not ship more softwaretech leadMeasure PR lead time and review capacity rather than lines of code, and set agent workloads accordingly.CTODo not base a productivity business case on lines of code generated, but on features delivered and lead time.senior developerExpect more review work to land on you because of agents; agree PR size limits in advance.Atlassian: AI time savings are disappearing into code reviewtech leadReview is the bottleneck, so add capacity or automation there.senior developerMore AI-generated code means more review, so reviewer time must be budgeted.AI code review is the next bottleneck: more code, fewer lines readtech leadThe review process needs redesigning, because the volume of code from agents exceeds human review capacity.senior developerYou have to decide what to review line by line and what to leave to automated checks.
Platform-level AI review and automatic approval
Copilot code review adds agent skills and MCP; impact dashboard gets ROI sectiontech leadSkills and MCP can be plugged into code review, and the dashboard's ROI data can be used to justify costs.CTOThe ROI view can help you judge whether Copilot licences are justified.Research: more AI agents are reviewing other AI agents' code on GitHubtech leadDecide which merges require human review even when the AI review has passed.QA testerDo not rely on AI review alone; build an independent testing layer.Copilot code review can now approve pull requeststech leadDecide whether a Copilot approval counts toward required reviews, and in which repositories it can be disabled.senior developerHave a human look over approved PRs too, especially for security and business logic.Copilot code review can approve pull requests without human review, off by defaulttech leadSet a policy on whether AI approval may be enabled and for which risk levels of change.senior developerThe role of human review is changing, but human approval remains mandatory for critical changes.Harness launches AI code review and rebuilds its Git repository for agent traffictech leadAI code review can help with the review bottleneck if false alarms are measured.GitHub Copilot adds cost and quality controls, and Code Review gains agent capabilitiestech leadSet a budget and an approval process for extra credits, and try out agentic Code Review.CTOCredit-based billing requires cost oversight, and the new controls can help with that.senior developerAgentic review can complement manual review, but it does not replace it.Copilot code review gets API support and a new default effort leveltech leadReview how PR review depth and cost change under the new default, and build the API into your workflow.Codex Auto-review goes free for ChatGPT users as GitHub changes Copilot review billing and Anthropic opens Claude Code Review betatech leadAI review is now available from three providers, so it is worth piloting them on cost and impact.CTOBecause of the organization-level billing settings for Copilot code review, the budget needs to be set in advance.QA testerAutomated review can be used to pre-screen pull requests, but it does not replace your own checks.AI code review: Atlassian splits it between two agents, JetBrains proposes a three-tier reviewtech leadDefine what the AI checks and what humans check, and tailor the review process accordingly.senior developerUse the AI reviewer for pre-screening, but keep logic and architecture checks for yourself.Anthropic makes Code Review in Claude Code generally availabletech leadTry it with one team and compare the bugs it finds with your current review results.senior developerAI review can work well as a first-pass filter, but final approval should stay with a human.
How reliable is AI review?
GhostCommit exposes a blind spot in AI code reviewtech leadAdd a supplementary check alongside AI review before merging.GitHub launches ReviewBench, an open benchmark for AI code reviewtech leadAn open yardstick is now available for comparing code review tools; it is worth using in purchasing decisions.QA testerThe benchmark helps evaluate how well AI reviewers detect defects.CTOAn independent benchmark can be used to check vendor claims before purchase.Meta study: two AI coding agents catch more bugs than one with a larger budgetQA testerCode should be checked by several independent agents rather than a single one.tech leadIn the review process, combining several smaller agents can be more cost-effective.PostgreSQL 19 delayed as AI review finds 40 bugssenior developerOn a large project, run an AI review before release as a supplement.tech leadAllow for the possibility that AI review can also push back the release calendar.AI code review finds bugs, delaying PostgreSQL 19senior developerTry AI review on critical codebases as an additional bug-finding layer.Greptile: DIY AI code reviewer costs $3 to $25 per pull request and still misses the key bugtech leadThe cost of an in-house review bot can be calculated per PR, but deeper bugs still need human review.
Investors bet on the control layer
AI coding boom shifts software companies' focus to testing and verificationQA testerThe value of testing is rising, so build your skills towards automation and verification of AI-written code.tech leadScale testing and verification capacity in proportion to where AI speeds up code writing.CodeRabbit raises $143 million at a $1.5 billion valuationtech leadAssess whether your team's pull request process needs a separate AI review layer, as the market is investing in this.CTOThe funding signals that reviewing AI code will become a separate cost line.Chinese AI code-review startup becomes a unicorn at a 10.8 billion yuan valuationstartup founderCode review is a market that attracts capital, but the field is crowded with competitors.CTOThe choice of AI review tools is widening, so it is worth starting vendor evaluation now.Blacksmith, an AI code-testing startup, reaches $1 billion valuationQA testerTesting and CI capacity are the bottleneck, so automation skills are becoming more valuable.startup founderTesting infrastructure is emerging as a market in its own right alongside AI coding.
Google search interest
Hungary
| Topic and search term | Weekly | Average |
|---|---|---|
| Code review has become the new bottleneckcode review | 0 | |
| The gain is lost in reviewpull request review | 0.9 | |
| Platform-level AI review and automatic approvalcopilot code review | 2 | |
| How reliable is AI review?AI code review | 0 | |
| Investors bet on the control layerAI startup investment | 0 |
World
| Topic and search term | Weekly | Average |
|---|---|---|
| Code review has become the new bottleneckcode review | 12.9 | |
| The gain is lost in reviewpull request review | 0 | |
| Platform-level AI review and automatic approvalai code review | 1 | |
| How reliable is AI review?ai code review accuracy | 0 | |
| Investors bet on the control layerai code verification startup | 0 |
Google Trends relative search index over the period studied, weekly average. A narrative and its sub-threads appear in a single comparison: 100 is the group's most-searched day, and the figures can only be compared within the group. The search term was chosen by machine analysis; it measures how much the term is searched, not how widespread the narrative is.
Who is affected and how
- tech lead
- It is advisable to introduce a review rule based on risk level. For low risk, AI review is sufficient, but for security, data, and infrastructure changes, human approval must be mandatory. Copilot's automatic PR approval is disabled by default and should remain so.
- senior developer
- The review burden falls on you. Recurring feedback should be codified into rules and checks to avoid commenting on the same issues repeatedly.
- QA tester
- Review and testing blur. It is worth building the validation of review-agent outputs and ReviewBench-style measurement as a core competency.
- CTO
- When selecting review tools, the rate of missed errors must be measured, not the number of comments. Costs per pull request and variable billing must be accounted for: Codex Auto-review is free for ChatGPT users, while Copilot code review is introducing new organizational billing options.
Hungary
The list contains no [HU]-marked stories, so the narrative has minimal domestic footprint during the period.
Weekly trend
Turning points
Copilot code review can now approve pull requestsCopilot code review can now approve pull requests, not just add comments. The feature is off by default. This raises the question of whether human approval is still needed.AI broke code review: experts disagree on what should replace itGergely Orosz and The New Stack state that AI generates more code than developers can follow, so the review process must adapt. Experts debate what will replace current practices.Anthropic makes Code Review in Claude Code generally availableOn October 15, 2026, Anthropic made Code Review generally available in Claude Code. As a result, the tool that generates code and the tool that checks it come from the same vendor.
Supporting stories
Agentic AI rewrites code review economics as Rootly drops its small-PR ruletech leadRules on PR size and review process need rethinking for the volume of change agents produce.senior developerSplit large AI PRs into smaller, stacked changes before review.QA testerReviews can rely on automated evidence (tests, checks) instead of line-by-line reading.AI coding boom shifts software companies' focus to testing and verificationQA testerThe value of testing is rising, so build your skills towards automation and verification of AI-written code.tech leadScale testing and verification capacity in proportion to where AI speeds up code writing.GhostCommit exposes a blind spot in AI code reviewtech leadAdd a supplementary check alongside AI review before merging.CodeRabbit raises $143 million at a $1.5 billion valuationtech leadAssess whether your team's pull request process needs a separate AI review layer, as the market is investing in this.CTOThe funding signals that reviewing AI code will become a separate cost line.Articles urge risk-tiered rules as AI-generated code review burden growstech leadIntroduce a risk-tiered review rule so that large AI-generated diffs do not slip through without being reviewed.senior developerReduce the review load with smaller units of change and automated checks.QA testerTests and acceptance criteria will play a more prominent role in verifying AI-generated code.Chinese AI code-review startup becomes a unicorn at a 10.8 billion yuan valuationstartup founderCode review is a market that attracts capital, but the field is crowded with competitors.CTOThe choice of AI review tools is widening, so it is worth starting vendor evaluation now.LinkedIn uses a multi-agent approach for AI code review at scaletech leadThe multi-agent review architecture can serve as a model, but it should first be tested on a small area of the codebase.senior developerAlongside automated review, human review remains the final gate.Research: more AI agents are reviewing other AI agents' code on GitHubtech leadDecide which merges require human review even when the AI review has passed.QA testerDo not rely on AI review alone; build an independent testing layer.Copilot code review can now approve pull requeststech leadDecide whether a Copilot approval counts toward required reviews, and in which repositories it can be disabled.senior developerHave a human look over approved PRs too, especially for security and business logic.Copilot code review can approve pull requests without human review, off by defaulttech leadSet a policy on whether AI approval may be enabled and for which risk levels of change.senior developerThe role of human review is changing, but human approval remains mandatory for critical changes.AI broke code review: experts disagree on what should replace ittech leadThe review process needs redesigning: what humans check, what machines check, and the upper limit on PR size.senior developerReview becomes a core competence, as checking for errors is harder than writing the code.QA testerAutomated test coverage becomes the main line of defense against AI-generated changes.Anthropic describes how agentic coding strains CI and how it scaled test impact analysisQA testerRunning only the affected tests, rather than the full suite, can be the solution.tech leadCI costs driven by agents can be planned for if tests are selected based on changes.GitHub launches ReviewBench, an open benchmark for AI code reviewtech leadAn open yardstick is now available for comparing code review tools; it is worth using in purchasing decisions.QA testerThe benchmark helps evaluate how well AI reviewers detect defects.CTOAn independent benchmark can be used to check vendor claims before purchase.Codex Auto-review goes free for ChatGPT users as GitHub changes Copilot review billing and Anthropic opens Claude Code Review betatech leadAI review is now available from three providers, so it is worth piloting them on cost and impact.CTOBecause of the organization-level billing settings for Copilot code review, the budget needs to be set in advance.QA testerAutomated review can be used to pre-screen pull requests, but it does not replace your own checks.AI-generated code is burning out senior engineers in code review, analysis sayssenior developerRecurring review feedback should be recorded as rules so that automation can check it.tech leadPlan how review load is shared and how AI reviewers are introduced before reviewers burn out.AI code review: Atlassian splits it between two agents, JetBrains proposes a three-tier reviewtech leadDefine what the AI checks and what humans check, and tailor the review process accordingly.senior developerUse the AI reviewer for pre-screening, but keep logic and architecture checks for yourself.Anthropic makes Code Review in Claude Code generally availabletech leadTry it with one team and compare the bugs it finds with your current review results.senior developerAI review can work well as a first-pass filter, but final approval should stay with a human.Blacksmith, an AI code-testing startup, reaches $1 billion valuationQA testerTesting and CI capacity are the bottleneck, so automation skills are becoming more valuable.startup founderTesting infrastructure is emerging as a market in its own right alongside AI coding.Greptile: DIY AI code reviewer costs $3 to $25 per pull request and still misses the key bugtech leadThe cost of an in-house review bot can be calculated per PR, but deeper bugs still need human review.AI code review is the next bottleneck: more code, fewer lines readtech leadThe review process needs redesigning, because the volume of code from agents exceeds human review capacity.senior developerYou have to decide what to review line by line and what to leave to automated checks.Qodo builds context into AI code review with Codextech leadAssess whether context-based review produces fewer false alarms than your current tool.
Machine analysis based on story titles and summaries, with a verification pass. The source of the facts is the article linked at each story.