---
title: "Discussion on \"From 99.2% to 84.8%: How I made my AI detector actually better by making it worse\" | Hashnode"
description: "Discussion on \"From 99.2% to 84.8%: How I made my AI detector actually better by making it worse\". In my last post, I talked about building an AI image detector without deep learning — pure signal processing, an SVM, and math borrowed from research papers. I ended that post admitting the tool strug"
url: "https://hashnode.com/posts/ai_image_detector_blog_2/69a1495620171868092fef6c"
---

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

Discussion

# [From 99.2% to 84.8%: How I made my AI detector actually better by making it worse](https://aidetector.hashnode.dev/ai_image_detector_blog_2?utm_source=hashnode&utm_medium=feed)

In my last post, I talked about building an AI image detector without deep learning — pure signal processing, an SVM, and math borrowed from research papers. I ended that post admitting the tool strug

Feb 27 · 6 min read · aidetector.hashnode.dev  
Read the article

## Written by

[Aman Chaturvedi](https://hashnode.com/@PrivacyAman)

@PrivacyAman

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

Fighting for privacy and technology to find AI generated images

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
