---
title: "Study: AI coding agents produce more code, not more software"
url: https://www.parallelquant.com/posts/study-ai-coding-agents-produce-more-code-not-more-software-9374e0
source_name: "Ars Technica"
source_url: https://arstechnica.com/ai/2026/10/ai-coding-agents-generate-more-code-but-not-more-software/
published: 2026-10-09T19:43:50.000Z
topics: ["research", "agents"]
publisher: "Parallel Quant"
---

# Study: AI coding agents produce more code, not more software

*2026-10-09 · Source: [Ars Technica](https://arstechnica.com/ai/2026/10/ai-coding-agents-generate-more-code-but-not-more-software/)*

A study found that coding efficiency gains from AI agents are absorbed by a human review bottleneck. Teams generate more code but do not ship proportionally more software.

**Why it matters:** It is empirical evidence that productivity from code generation is capped by verification, not writing. That strengthens the case for investing in automated review and testing, and fits with multi-agent bug-finding tools targeting that bottleneck. It also tempers headline claims about AI-driven engineering throughput.

**Topics:** research, agents

---
Read the original: https://arstechnica.com/ai/2026/10/ai-coding-agents-generate-more-code-but-not-more-software/
Canonical: https://www.parallelquant.com/posts/study-ai-coding-agents-produce-more-code-not-more-software-9374e0
Published by Parallel Quant — https://www.parallelquant.com