Edition of Wednesday 11 November 2026

Is AI code rotting open source? 450 repos, 12 months

Comments on “Is AI code rotting open source? 450 repos, 12 months”

Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked

Coderot, a code-analysis vendor, scanned 450 popular GitHub repos and found that AI-signed merged PRs averaged 11.5% in September 2026, up from 2.3% a year earlier. The median repo sits at 1.5%, and signatures only count traces tools leave, so these are floors. In 22 AI-heavy TypeScript and C# repos, complexity per line rose a median 4.8% against 3.0% elsewhere, concentrated in fast-growing codebases. AI PRs run about 2.4 times larger. The figures come from the vendor's own tool and are unverified, and the VS Code case is uncorroborated.

Read the original at coderot.ai. The judges saw only its text, with the site, the author and link domains removed; they are shown again after ranking.

Why it ranked

No. 7 of 143 with a score of 0.72. The panel rated it highest on substantiated (0.90) and reader interest (0.85), lowest on significance (0.70). Placed No. 7, No. 10 by score alone: the placement terms moved other stories past it.

AI judgeJudge 2claude-haiku-5-5
Large-sample measurement study of AI-signed commits across 450 popular repos with transparent methodology; highly relevant to builders, though share is a floor and correlational claims are caveated.
AI judgeJudge 3gpt-5.6-luna
Large, unusually detailed empirical study of AI-assisted open-source development, with explicit sampling, metrics, limitations, repo-level evidence, and reproducible methodology. Its tool promotion is secondary; conclusions are notably cautious.

Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.

Score

0.72total under rubric v2.3

Judges’ mean answer per question, on a scale from 0 to 1. Questions marked minus count against the score.
  • Significance0.70
  • Reader interest0.85
  • Audience breadth0.81
  • Practical relevance0.75
  • Substantiated0.90
  • Technical depth0.82
  • New information0.84
  • Counts against: Promotional0.20
  • Counts against: Overclaim0.25

Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).

QuestionJudge 1Judge 2Judge 3MeanSpreadWeightContribution (weight times mean)
SignificanceHow much does this change what people in tech know or can do?0.730.600.780.700.18+0.24+0.169
Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on?0.800.850.900.850.10+0.26+0.221
Audience breadthHow many people who build technology does it concern?0.860.750.820.810.11+0.13+0.105
Practical relevanceHow useful is it to someone who builds software or tech products?0.800.700.740.750.10+0.22+0.164
SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations.0.990.800.900.900.19+0.05+0.045
Technical depthHow technically substantive is it?0.970.700.800.820.28+0.05+0.041
New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)?0.830.700.990.840.29+0.05+0.042
PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional.0.060.250.300.200.24−0.19−0.039
OverclaimDo the headline or conclusions claim more than the body supports?0.340.300.120.250.22−0.11−0.028
Sum of contributions+0.721
Claim-check penalty0.000
Total, as published0.721

Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.

The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.

Ranked #7 on Rubric, 11 Nov 2026
<a href="https://rubric.hrih.si/editions/2026-11-11/7/"><img src="https://rubric.hrih.si/badge/2026-11-11/7.svg" alt="Ranked #7 on Rubric, 11 Nov 2026" width="238" height="28"></a>

Claim check

Claim check: 6 no source given · 2 couldn’t check. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.

  1. No source givenNumber

    Across 450 popular GitHub repos, the average share of merged PRs signed by an AI agent was 11.5% in September 2026, up from 2.3% in October 2025.

    The article gives no link or citation for it.

    The article says

    “the average one of 450 of GitHub’s most popular repos had 11.5% of its merged pull requests signed by an AI agent. In October 2025 it was 2.3%”

    The source says

    No source to compare with.

  2. No source givenNumber

    In 279 of the 388 repos with AI-signed PRs, Claude Code is the tool that signs the most of them.

    The article gives no link or citation for it.

    The article says

    “279of the 388 repos with AI-signed PRs have Claude Code as the tool that signs the most”

    The source says

    No source to compare with.

  3. No source givenNumber

    Complexity per 1,000 lines rose a median 4.8% in the 22 AI-heavy TypeScript and C# repos versus 3.0% in the other 63.

    The article gives no link or citation for it.

    The article says

    “4.8%typical change in code complexity over the year in the 22 AI-heavy TypeScript and C# repos; +3.0% in the other 63”

    The source says

    No source to compare with.

  4. No source givenCausal conclusion

    AI-heavy repos' extra complexity appeared only where the codebase was also growing fast; among slow-growing repos the AI-heavy ones rose less (1.7% vs 2.5%).

    The article gives no link or citation for it.

    The article says

    “AI-heavy repos got a little more complex than the rest, all of it where they were also growing fast”

    The source says

    No source to compare with.

  5. No source givenNumber

    In 55 of 61 repos with enough PRs, the median AI PR is bigger than the median other PR, typically 2.4 times as big.

    The article gives no link or citation for it.

    The article says

    “In 55 of the 61 repos with at least 10 AI PRs and 10 others in the last six months, the median AI PR is bigger than the median other…”

    The source says

    No source to compare with.

  6. Couldn’t checkDate

    VS Code's git.addAICoAuthor setting added Co-authored-by: Copilot to commits by default from 15 April 2026 until VS Code 1.119, even when no AI was used.

    The source was unreachable, paywalled or too long, or the two checkers disagreed.

    The article says

    “From 15 April 2026 until VS Code 1.119, its git.addAICoAuthor setting added Co-authored-by: Copilot to commits by default, even when no AI was used”

    The source says

    No passage could be compared.

    Source cited by the article: theregister.com/2026/05/04/microsoft_reverses_ai_credit_grab

  7. No source givenNumber

    AI-signed PRs were fixed within 30 days only slightly more often than other PRs within each repo, a median of 1 point.

    The article gives no link or citation for it.

    The article says

    “Within a repo, AI PRs are fixed within 30 days only a little more often than the rest, a median 1 point”

    The source says

    No source to compare with.

  8. Couldn’t checkNumber

    Agents took over VS Code's merged PRs in August 2026, with 71% of PRs in the last three months carrying an AI signal, while its complexity per 1,000 lines rose from 182 to 201.

    The source was unreachable, paywalled or too long, or the two checkers disagreed.

    The article says

    “Agents took over in August 2026. Since then, 71% of merged PRs carry an AI signal”

    The source says

    No passage could be compared.

    Source cited by the article: app.coderot.ai/demo/microsoft/vscode

Also covered by

No other coverage was clustered with this story.

Elsewhere

Elsewhere: HN 1 pts, not on front page

Discussion on Hacker News

Recorded for context only, checked 11 Nov 2026, 05:47 UTC. It played no part in the ranking.

Panel for this edition

Four AI agents ranked this edition: three judges scored every story blind to its site and author, and a fourth compared the top stories’ claims with the sources they cite. Summaries are written by AI.

Model versions are pinned for every edition. How ranking works describes rubric v2.3.

Comments

Readers’ comments on this story have a page of their own, ranked by a public rubric rather than by votes.

Comments on “Is AI code rotting open source? 450 repos, 12 months”

More on Rubric

From the last 7 editions, chosen by shared topic, score and date.

  1. Score 0.75

    llama.cpp v1.0