---
title: "Discussion on \"AI Agents Don't Fail at Code. They Fail at Learning.\" | Hashnode"
description: "Discussion on \"AI Agents Don't Fail at Code. They Fail at Learning.\". SWE-CI tested 18 AI models across 71 consecutive commits. Most broke something on commit 47 they'd already broken on commit 1. That's not an intelligence problem. That's a learning system that isn't learning."
url: "https://hashnode.com/posts/ai-agents-don-t-fail-at-code-they-fail-at-learning/69f646c20ab374db9927b8c0"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [AI Agents Don't Fail at Code. They Fail at Learning.](https://ynagarjuna.hashnode.dev/ai-agents-don-t-fail-at-code-they-fail-at-learning?utm_source=hashnode&utm_medium=feed)

SWE-CI tested 18 AI models across 71 consecutive commits. Most broke something on commit 47 they'd already broken on commit 1. That's not an intelligence problem. That's a learning system that isn't learning.

May 2 · 10 min read · ynagarjuna.hashnode.dev  
Read the article

-   [#ai](https://hashnode.com/tag/ai)
-   [#maintainability](https://hashnode.com/tag/maintainability)
-   [#engineering-leadership](https://hashnode.com/tag/engineering-leadership)
-   [#code-quality](https://hashnode.com/tag/code-quality)

## Written by

[Nagarjuna Yelisetty](https://hashnode.com/@ynagarjuna)

@ynagarjuna · Bengaluru, India

  
[View profile](https://hashnode.com/@ynagarjuna)

## Responses

Markdown supported · ⌘↵ to post

No responses yet.