原始对话

AI-Generated Code Is Already Competing With Human Code — Daksh Gupta, Greptile

AI Engineer · 2026年9月27日 · 12:40

Daksh Gupta discusses AI-generated pull requests, how their revert rates compare with human-written changes, and differences in the types of errors coding agents produce.

Daksh Gupta

在YouTube上观看

AI Engineer

一目了然

Rapid growth of AI-generated pull requests[1]阅读支持性片段
Revert rates show minimal difference between human and agent PRs[2]阅读支持性片段
AI coding agents exhibit distinct qualitative failure patterns[3]阅读支持性片段

关键时刻3

简短、标注来源的段落，并附有可验证上下文。完整对话保留在其发布者处。

4:34

AI产品

Rapid growth of AI-generated pull requests

原始摘录

And I came to the result that about a quarter of all the poll requests that Graal was reviewing in any given month were completely or at least largely generated by AI. Then I backtracked this data to the last 12 months and it turns out that this number is growing really fast. In fact, early last year, fewer than 1% of pull requests had any evidence of being completely AI generated.

Daksh Gupta·2026年9月27日

此片段未附更多上下文，请阅读原始访谈。

5:55

AI产品

Revert rates show minimal difference between human and agent PRs

原始摘录

Codex PR is reverted about one out of every thousand poll requests. Devon once every three and a half times every thousand poll requests. Humans right in the middle at about two and a half. So there doesn't seem to be very big difference between the rate at which poll requests were reverted from people versus agents in my study.

Daksh Gupta·2026年9月27日

此片段未附更多上下文，请阅读原始访谈。

9:16

AI产品

AI coding agents exhibit distinct qualitative failure patterns

原始摘录

Claude is one and a half times more likely to produce a SQL injection error than humans. Devon is about half as likely as humans to produce a off bypass issue. I found it very interesting that there was this much variation in how these agents were performing and how different their failure modes were from humans.

Daksh Gupta·2026年9月27日

此片段未附更多上下文，请阅读原始访谈。

来源与研究方法

这些观点均关联原始来源。转述已明确标注，不作为逐字原话展示。

打开转录或来源材料报告问题