𝗪𝗵𝘆 𝗔𝗜-𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗲𝗱 𝗖𝗼𝗱𝗲 𝗢𝗳𝘁𝗲𝗻 𝗟𝗼𝗼𝗸𝘀 “𝗖𝗼𝗺𝗽𝗹𝗲𝘁𝗲” — 𝗯𝘂𝘁 𝗜𝘀𝗻’𝘁—𝗮𝗻𝗱 𝘄𝗵𝘆 𝗜 𝗯𝘂𝗶𝗹𝘁 𝗔𝗜-𝗦𝗟𝗢𝗣 𝗗𝗲𝘁𝗲𝗰𝘁𝗼𝗿

#opensource #codequality #devsecops #programming

A few days ago, I published a repository called:
*“HRPO-X v1.0.1 – Hybrid Reasoning Policy Optimization Framework.”
*
I genuinely believed it was solid work:
▪️Paper-inspired architecture
▪️Clean folder structure
▪️Configs in place
▪️Interfaces and classes defined
▪️Even internal audit checks passing
Then I saw this comment:
“𝑨𝒔 𝒆𝒙𝒑𝒆𝒄𝒕𝒆𝒅 — 𝒂𝒏 𝑨𝑰 𝒔𝒍𝒐𝒑 𝒓𝒆𝒑𝒐 𝒃𝒖𝒊𝒍𝒕 𝒐𝒏 𝒉𝒂𝒍𝒍𝒖𝒄𝒊𝒏𝒂𝒕𝒊𝒐𝒏𝒔.”

At first, I ignored it.
Then I re-read the code.
They were right.

𝑾𝒉𝒚 𝒕𝒉𝒊𝒔 𝒉𝒂𝒑𝒑𝒆𝒏𝒔
The issue wasn’t intent or effort.
It was density.

AI tools are great at producing structurally correct artifacts:
▪️Proper folder hierarchies
▪️Configuration files
▪️Class and interface definitions
▪️Clean pipelines and entry points

Most linters, CI checks, and even internal audits focus on exactly these signals.

But AI often fails at something more subtle:
👉 𝑴𝒆𝒂𝒏𝒊𝒏𝒈𝒇𝒖𝒍 𝒊𝒎𝒑𝒍𝒆𝒎𝒆𝒏𝒕𝒂𝒕𝒊𝒐𝒏 𝒅𝒆𝒏𝒔𝒊𝒕𝒚

You end up with code that is:
▪️Empty functions
▪️Minimal logic
▪️Documentation that outweighs implementation.

𝑻𝒉𝒂𝒕’𝒔 𝒘𝒉𝒂𝒕 𝑰 𝒄𝒂𝒍𝒍 𝑨𝑰 𝑺𝒍𝒐𝒑.

𝗪𝗵𝘆 𝗲𝘅𝗶𝘀𝘁𝗶𝗻𝗴 𝘁𝗼𝗼𝗹𝘀 𝗺𝗶𝘀𝘀 𝗶𝘁
Traditional tools ask:
▪️Does it compile?
▪️Is the structure valid?

They rarely ask:
▪️How much real logic is here?
▪️Is the documentation proportional to the code?
That gap is where AI-generated slop thrives.

𝗦𝗼 𝗜 𝗯𝘂𝗶𝗹𝘁 𝗔𝗜-𝗦𝗟𝗢𝗣 𝗗𝗲𝘁𝗲𝗰𝘁𝗼𝗿
I built it to measure the gap between appearance and substance.
It statically analyzes Python code using signals like:
▪️Logic Density Ratio (LDR)
▪️Buzzword Inflation
▪️Unused dependencies (DDC)
▪️Common AI-generated patterns
These are combined into a single Deficit Score (0–100)
that reflects how hollow a codebase might be.
This isn’t about blaming AI or developers.

𝑾𝒉𝒚 𝒕𝒉𝒊𝒔 𝒊𝒔 𝒖𝒔𝒆𝒇𝒖𝒍

This tool isn’t about blaming:
▪️AI
▪️No-code or Low-code Developers

It’s for anyone who has looked at a repository and thought:
“This looks impressive… but something feels off.”

AI-SLOP Detector gives language and metrics to that intuition.
It helps reviewers, educators, and teams explain why a codebase feels wrong — even when everything appears structurally correct.

𝗔 𝗳𝗶𝗻𝗮𝗹 𝗻𝗼𝘁𝗲

This project came from embarrassment, frustration, and curiosity — but it led to a clearer understanding of a growing problem in the AI era.

If this resonates with your experience reviewing AI-generated code, I’d love to hear how you’ve been dealing with it.

👇 First comment AI-SLOP Detector Repo(MIT)