Generator Metadata X-Ray
Parses PNG tEXt/iTXt/zTXt chunks byte-by-byte. A leftover parameters or prompt field from Stable Diffusion, ComfyUI or Midjourney is smoking-gun evidence — weighted highest in the score.
FORENSIC LAB · IN-BROWSER · NO UPLOAD
AI Photo Detector — Check If a Photo Is AI-Generated
Drop an image into the lab. Our AI photo detector reads generator metadata, camera EXIF, C2PA content credentials and pixel statistics — then hands you an evidence-based probability report card. Free, instant, and 100% private: analysis runs in your browser.
SPECIMEN INTAKE01
Drag & drop a photo here
or browse files — PNG · JPEG · WebP · max 20 MB
Human verification unlocks the optional cloud check.
STANDBY — drop an image to arm the scanner
AI PROBABILITY REPORT CARD02
No specimen analyzed yet.
The evidence report renders here after your first scan.
—
EVIDENCE LOG
⚖ This score is a probability, not proof. Metadata can be stripped and heuristics can misread heavily edited photos. Use it as one signal among several.
AI Image Detector Features, Built Like a Forensics Kit
Four independent evidence channels — not one black-box guess.
Camera EXIF Cross-Check
A hand-written EXIF parser reads Make, Model, Software, timestamps and GPS. A complete camera fingerprint leans real; a wiped or AI-tool-signed one leans synthetic.
C2PA Credential Scan
Scans raw bytes for JUMBF/C2PA provenance blocks — the signed "birth certificate" that DALL-E and Adobe Firefly embed. If credentials exist, they appear as a dedicated line of evidence.
Pixel-Statistics Heuristics
Canvas-level analysis of noise uniformity, local variance and histogram smoothness. Clearly labeled heuristic and weighted lowest — honest signals, no overclaiming.
Private by Architecture
The pipeline runs in your browser — your photo never touches our servers unless you opt into the cloud check. Nothing stored, nothing logged, nothing to leak.
Shareable Report Card
Every scan produces a downloadable PNG report card: gauge, verdict, and the full evidence log with weights — ready to drop into a chat, a doc or a dispute.
What Is an AI Generated Photo Detector?
An AI generated photo detector is a forensic tool that estimates whether an image came out of a camera or out of a model like Midjourney, DALL-E or Stable Diffusion. Instead of trusting a single neural-network guess, this AI photo detector works like an evidence lab: it stacks independent signals — embedded generator parameters, camera EXIF fingerprints, C2PA content credentials and pixel-level statistics — and weighs each one according to how conclusive it really is.
Hard evidence (a prompt literally stored inside the file) can push the score to near-certainty. Soft evidence (unusually uniform noise) only nudges it. The report card shows you exactly which is which, so you can defend the verdict — not just repeat it.
EVIDENCE WEIGHTING
Generator metadata (PNG chunks)CONCLUSIVE
Software tag names AI toolSTRONG
C2PA / JUMBF credentialsSTRONG
Camera EXIF completenessMODERATE
Pixel-statistics heuristicsWEAK · HEURISTIC
Optional cloud model checkMODERATE
How to Detect AI Photos in 3 Steps
From suspicion to evidence in under ten seconds.
-
01
Drop in the photo
Drag any PNG, JPEG or WebP into the intake panel — or click to browse. The file is read locally; nothing is uploaded.
-
02
Hit Analyze & scan
Hit Analyze Image and watch the four-channel AI photo detector pipeline sweep the file in real time. A quick human check unlocks the optional cloud model on top.
-
03
Read the report card
Get a 0–100 AI probability, a plain-English verdict and a line-by-line evidence log. Download the card if you need a receipt.
Why Use This AI Photo Checker?
Evidence, not vibes
Most detectors return a bare percentage. Ours shows its work: every signal, its direction, and its weight. If the score says 92, you will know why it says 92.
Radically private
Journalists, moderators and teachers check sensitive images. In-browser analysis means the photo you check never becomes someone else's data.
Honest about limits
Heuristics are labeled heuristics. Missing metadata lowers confidence and the report says so. An AI photo checker you can cite is better than one that bluffs.
Fast and free
No signup, no credits, no watermarks. This free AI photo detector is hand-written JavaScript that runs in milliseconds on your own device.
Free Where It Costs Us Nothing. Paid Where It Doesn't.
On-device analysis is unlimited and free, forever. Cloud recheck calls a paid third-party engine, so heavy use is a Pro feature.
Free
$0 no signup, no credits
- Unlimited on-device scans (4 evidence channels)
- Full evidence log & probability score
- Downloadable report card
- Cloud recheck: 3 per day
Most Popular
Pro
$4.99/mo or $47.90/yr — save 20%
- Everything in Free
- 200 cloud rechecks per month
- Batch scan up to 20 images
- Priority recheck queue
- Scan history, 30 days
AI Photo Detector FAQ
What is an AI photo detector and how does it work?
An AI photo detector analyzes an image for traces left by generators such as Midjourney, DALL-E or Stable Diffusion. This tool inspects PNG text chunks for embedded prompts, parses JPEG EXIF camera fields, scans for C2PA content credentials, and runs pixel-statistics heuristics. Each finding is weighted into a 0–100 AI probability score.
How accurate is AI image detection?
No detector is 100% accurate. Embedded generator metadata is near-conclusive evidence, while pixel heuristics are only statistical hints. That is why the report card lists every piece of evidence with its weight and states the result as a probability, not proof.
Is this photo AI generated — can I check for free?
Yes. Upload any photo and you get a free AI probability report card in seconds. No account, no watermark, no scan limit — the analysis runs locally in your browser.
Does the AI photo checker upload my images to a server?
No. The core forensic pipeline — metadata parsing, C2PA scan and pixel analysis — runs entirely in your browser. Your photo only leaves your device if you explicitly tick the optional cloud check.
Which AI image generators can be detected?
Metadata signatures cover Stable Diffusion, ComfyUI, AUTOMATIC1111, Midjourney, DALL-E, Flux, SDXL, NovelAI, InvokeAI, Adobe Firefly and Leonardo AI, among others. Images with stripped metadata fall back to the C2PA scan and pixel-statistics analysis.
What are C2PA content credentials?
C2PA (Coalition for Content Provenance and Authenticity) is an open standard that embeds a signed history inside an image file. Tools like Adobe Firefly and DALL-E attach these credentials. Finding a C2PA/JUMBF block is strong provenance evidence, and the report card lists it as a separate finding.
Can AI detection be fooled?
Yes. Screenshots, re-encoding and social-media compression strip metadata, and heavy editing changes pixel statistics. When strong evidence is missing, the score leans on weaker heuristics and the report clearly labels the lower confidence.
Which image formats are supported?
PNG, JPEG and WebP up to 20 MB. PNG gets the deepest inspection (generator parameters often hide in tEXt/iTXt chunks), JPEG gets full EXIF parsing, and every format goes through the C2PA scan and pixel heuristics.