lyq9797

lyq9797 / aigc_web

Public

The AI text detection system assists users in conducting fine-grained detection of AI-generated content within mixed texts

17
11
85% credibility
Found May 25, 2026 at 17 stars 4x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

This is a web-based tool that helps you determine whether a piece of text was written by a human or generated by artificial intelligence. Users create an account, paste text or upload documents, and the system analyzes each word and sentence to identify patterns typical of AI-generated content. The results are displayed with color-coded highlighting, showing exactly which parts of the text appear to be AI-written. Users can save their analysis history and export results for future reference.

How It Works

1
🌐 You discover the tool online

You find a website that can tell you if text was written by a human or generated by AI.

2
📝 You create your account

You sign up with a username and password to access the detection features.

3
✍️ You paste your text or upload a document

You enter the text you want to check, or upload a file like a Word document or plain text file.

4
🔍 The system analyzes your text

The tool examines every word and sentence, looking for patterns typical of AI-generated content.

5
📊 You see the results highlighted

Each word and sentence is color-coded to show which parts look human-written and which look AI-generated.

6
You save or share your results
📥
Export results

Download your analysis as a text file or JSON for your records.

📜
View history

Browse through all your previous text analyses anytime.

✅ You know the truth about your text

Now you can confidently tell whether content was written by a human or created by artificial intelligence.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is aigc_web?

AIGC Web is an AI-generated text detection system that identifies whether content was produced by AI or humans. It analyzes mixed texts at two levels: word-level token classification and sentence-level classification, flagging where the text switches from human-written to AI-generated. The system runs as a web service with a Python backend and JavaScript frontend, handling user authentication, text input, file uploads (.txt, .doc, .docx), and storing detection history in SQLite. Users paste or upload text, get highlighted results showing which words and sentences are likely AI-generated, and can export findings as JSON or plain text.

Why is it gaining traction?

The dual-granularity approach sets this apart from simple binary classifiers. Instead of just labeling text as "AI or human," it pinpoints the exact transition point where content shifts between sources. The web interface makes it accessible without command-line work, and the file upload support handles batch processing. The project also includes performance test scripts and security scan reports, suggesting attention to production readiness.

Who should use this?

Content moderators checking for AI-generated submissions would find the transition-point detection useful. Educators verifying originality in student work could benefit from sentence-level analysis. Researchers experimenting with AI text detection methods might use it as a baseline or starting point for their own models. Teams needing a self-hosted solution rather than third-party APIs could deploy this internally.

Verdict

This is an early-stage research project with 17 stars and a credibility score of 0.85%, indicating minimal community validation. The code shows promise with solid authentication practices and multi-granularity detection, but hardcoded paths, placeholder model addresses, and Windows-specific dependencies suggest it was built for a specific environment rather than general use. If you need production-ready AI text detection today, look elsewhere. If you're researching detection methods or want to build on this architecture, the source provides a reasonable foundation with the caveat that deployment will require cleanup.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.