Skip to content
DevLab

AI Image Detector — Metadata Inspector for ChatGPT / Midjourney / SD / ComfyUI

Drop an image to determine from metadata whether it was AI-generated. Analyzes embedded metadata (C2PA Content Credentials, PNG textChunks, EXIF Software, XMP) from major AI tools including ChatGPT (DALL·E 3), Sora, Adobe Firefly, Midjourney v7, Google Imagen 3, Microsoft Designer, Stable Diffusion, ComfyUI, NovelAI, InvokeAI, Leonardo.AI, Krea, Flux, and Apple Image Playground. Fully processed in-browser; images are never sent to a server.

100% Free No signup Browser-only No logs / DB 5 languages Dark mode
🔍

Drop an image or click to select

PNG / JPEG / WebP / AVIF · 100% browser-only

▶ 📚 Supported AI tools (16+)
ChatGPT (DALL·E 3 / GPT-image-1) — C2PA "OpenAI"
OpenAI Sora — C2PA video manifest
Adobe Firefly — C2PA "Adobe Firefly"
Microsoft Designer / Copilot — C2PA + EXIF Software
Google Imagen 3 / Nano Banana — SynthID + C2PA
Midjourney v6 / v7 — C2PA (since 2026), no PNG chunks
Stable Diffusion (AUTOMATIC1111 / Forge) — PNG `parameters` chunk
ComfyUI — PNG `prompt` + `workflow` zTXt chunks
NovelAI — PNG `Software: NovelAI` + `Comment` JSON
InvokeAI — PNG `invokeai_metadata` chunk
Fooocus — PNG `parameters` (A1111 互換)
Leonardo.AI — XMP / EXIF Software
Krea AI — XMP "Krea"
Black Forest Labs Flux — XMP / PNG metadata (一部)
Apple Image Playground — EXIF Software "Image Playground"
Pixel Magic Editor / Magic Eraser — XMP "Generative AI"

❓ Frequently Asked Questions

When detection fails?
Returns "no metadata". Screenshots, stripped EXIF, pre-2026 Midjourney, or edited images leave no metadata — but "no metadata" does NOT mean "not AI"
100% accurate?
When detected = factual (metadata literally says so). When not detected = unknown. High true-positive, but many false negatives possible (stripped metadata)
Privacy?
Fully in-browser. Image never leaves your device. Safe for confidential images
What is C2PA?
Industry standard from CAI (Adobe / OpenAI / Microsoft / BBC / Sony / Nikon) embedding cryptographically-signed content provenance into images
Can I see SD / ComfyUI prompts?
Yes. AUTOMATIC1111 / Forge / Fooocus write full prompt / model / seed / sampler / cfg to PNG `parameters` chunk. ComfyUI writes entire node graph as JSON to `workflow` chunk
🐛 Found a bug or issue with this tool?

Free to use, no signup. Even just the steps to reproduce are helpful. Reports go directly to the operator and help us fix issues.

* Browser info (UA / screen / language / URL) is sent automatically to help reproduce the issue