AI Image Detection
Identifies forensic artifacts commonly left by AI image generators, including Midjourney, DALL-E, Stable Diffusion, FLUX, and other generative models.
Free & Instant Analysis
Fake Image
Detector
Expose manipulated and AI-generated images in seconds with a four-layer forensic pipeline: AI detection, metadata analysis, Error Level Analysis and watermark scanning.
File too large. Maximum size is 10 MB.
- Privacy Protected
- No Signup Required
Capabilities
Everything you need to verify an image
A complete forensic toolkit: free, private, and fast enough for everyday fact-checking.
Metadata Analysis
Extracts EXIF metadata, embedded headers, software signatures, timestamps, GPS information, and other available metadata for forensic analysis.
Error Level Analysis (ELA)
Performs Error Level Analysis by recompressing images to reveal localized compression inconsistencies that may indicate image manipulation.
Privacy-First Processing
Every uploaded image is analyzed entirely in memory and is never permanently stored on the server, ensuring maximum privacy.
No Registration Required
Upload an image instantly without creating an account, subscribing, or providing personal information.
Ultra-Fast Analysis
Optimized processing delivers forensic results within seconds for a smooth user experience.
Broad Format Support
Supports the most common image formats (JPG, PNG, BMP, TIFF, and WebP) without requiring any preprocessing.
Advanced Forensic Reports
Provides easy-to-understand forensic reports highlighting potential manipulation indicators and authenticity signals.
Detection Pipeline
Four forensic layers, one verdict
Every image passes through independent forensic layers that each look for different traces of manipulation.
-
Layer 01
AI Deepfake Engine
Uses deep-learning classifiers to identify spatial frequency patterns and synthetic artifacts commonly found in AI-generated imagery.
-
Layer 02
Metadata Analysis
Examines EXIF metadata, software identifiers, compression history, and editing footprints.
-
Layer 03
Error Level Analysis
Highlights localized compression inconsistencies and pixel-level modifications through Error Level Analysis.
-
Layer 04
Watermark Detection
Detects supported invisible AI watermarks, steganographic markers, and embedded authenticity signals where available.
From the Blog
Resources & Insights
Stay ahead of evolving AI image generation techniques, digital forensics, and image authenticity detection.
July 20, 2026
14 min read
How JPEG Compression Works, and Why It Betrays Edited Photos
JPEG does not just shrink a photo, it rewrites it in 8x8 blocks and rounds away detail in a pattern you can measure. This guide explains the five steps behind every save, why an edited region ends up with a different compression history than its surroundings, and how to read that difference without jumping to conclusions.
Read article
July 19, 2026
13 min read
How to Verify Disaster Photos Before You Share or Donate
After every major disaster, the same three things flood social media: real photos from the wrong event, AI-generated scenes that never happened, and donation links attached to both. This guide covers four checks that take about two minutes, including reverse image search, metadata, Error Level Analysis, and vetting the charity separately from the photo. You will also learn which popular AI tells stopped working, and why a missing EXIF field proves nothing.
Read article
July 18, 2026
10 min read
C2PA vs SynthID: What Each One Really Proves About an Image
C2PA and SynthID both help you spot AI-generated images, but they work in completely different ways. This guide breaks down what Content Credentials and Google's invisible watermark each prove, where both quietly fail, and how to check an image in under a minute.
Read article
July 17, 2026
11 min read
Can ELA Detect AI Images? I Tested 10 on My Own Scanner, and Here's Where It Fails
Error Level Analysis is great at catching edited photos, but can ELA detect AI images? I ran 10 AI-generated pictures through the scanner on this site and logged every verdict. Here's what it caught, what it missed, and the layer that actually did the work.
Read article
July 16, 2026
10 min read
How to Spot Fake Product Photos When Shopping Online
Fake product photos are everywhere now, from lightly edited stock shots to fully AI-generated scenes that never existed. This guide shows you how to spot them fast, using free tools and a handful of visual checks, so you know what you're really buying before you spend a cent.
Read article
July 15, 2026
11 min read
SynthID Image Watermarks Explained: What They Catch and What They Miss
Google has hidden an invisible marker inside more than 100 billion AI-generated images and videos, and since May 2026 it rides inside pictures from ChatGPT as well. This guide explains how the SynthID image watermark works, how to check any picture for it in under a minute, and the one result almost everyone reads backwards.
Read article
FAQ
Frequently asked questions
Quick answers about privacy, usage, and accuracy.
No. Your image is analysed entirely in memory, is never written to disk, and is discarded the moment your results are generated. We do not keep, share, or reuse uploaded images for any purpose, and they are never used to train our models.
Drag your image into the upload box at the top of this page, or click browse to choose a file, then press Scan Now. The analysis runs immediately and shows your original image beside its ELA map, the metadata verdict, and any AI watermark or provenance flags.
JPG, PNG, BMP, TIFF, and WebP images up to 10 MB. Formats such as HEIC or AVIF are not supported yet, so convert those to JPG or PNG first. Whenever possible, scan the most original file you have, since every conversion is itself an edit.
Four independent layers: AI detection with machine-learning classifiers, metadata analysis of the file's EXIF data, Error Level Analysis of its compression history, and watermark and provenance detection including C2PA Content Credentials. Each looks for a different kind of evidence, so together they are far more reliable than any single technique.
Automated image forensics is probabilistic. Combining four independent layers makes results considerably more reliable than any single technique, but both false positives and false negatives happen. Treat results as strong signals to inform your own judgement, not conclusive proof.