PROOF
5 physical stages trace how a camera sensor converts light into pixels.
Physics, behavior, and metadata signals reveal whether a photo is original or modified.
Built on a dataset of real devices to learn unique camera fingerprints.
Proof extracts 80+ signals across three groups: physics, behavior, and metadata. Physics signals measure what the camera sensor leaves in every pixel. Behavior signals detect patterns from AI editors, social platforms, and processing pipelines. Metadata signals check file structure, compression tables, and encoding fingerprints. Our signals measure each stage independently.
If any stage is missing or altered, our signals detect it.
Every image is scored by five independent detectors:
Each detector weighs the signals that matter most for that specific image and shows exactly which signals influenced its answer.
AI generators synthesize pixels directly from neural networks. They skip the entire camera pipeline. No sensor noise, no Bayer pattern, no PRNU fingerprint, no hardware JPEG encoder. Our 80+ signals across physics, behavior, and metadata measure what should be there from a real camera. When it's missing, we know.
Most AI detectors use neural networks trained to recognize what AI output "looks like." They break every time a new generator launches. Proof measures the physics fingerprint the camera pipeline leaves in every pixel. Real cameras produce signals that are extremely difficult to replicate, regardless of what AI tool was used. Every result shows exactly which signals drove the verdict. No black box.
All uploaded images are automatically deleted after 24 hours. Your data is never sold or shared. If you need an image removed sooner, contact contactus@proofme.ai.
Download the App