ScrubAI.
Deep Verification Pipeline / Client-Side Wasm

C2PA Content Credentials
Verification Scanner

Verify whether an image was created by AI or modified using tools that attach cryptographically signed Content Credentials. This tool executes a sandbox WebAssembly compiler completely inside your browser to inspect binary structures.

🔍

Drop image here to scan

JPEG, PNG, WEBP, AVIF up to 15MB

Want to verify instantly with a sample?

Load a pre-configured AI image containing valid C2PA credentials.

Verification Briefing

Awaiting Inspection

Upload a file on the left or load our sample to run the deep signature verification.

Understanding Provenance Infrastructure

The C2PA Technical Briefing

01What is C2PA?

C2PA (Coalition for Content Provenance and Authenticity) is an open-source technical standard established by a consortium of tech giants, social platforms, and media publishers. It represents a standardized framework designed to verify the history and origin (provenance) of digital media assets to prove asset authenticity.

Unlike traditional EXIF camera metadata, which is stored in simple headers and easily modified, C2PA binds cryptographically secure signatures directly to image binary structures using standard JUMBF blocks.

02How It Works

When an image is captured with a C2PA-enabled camera or generated using an AI engine (like ChatGPT/DALL-E 3), the creator system generates a cryptographically signed document called an Assertion Manifest.

This manifest contains detailed claims about the creation tool, coordinates, timestamps, and edit actions. The manifest block is hashed and secured via digital signatures using public-key cryptography (e.g. RSA-PSS or Ed25519). If even a single pixel is subsequently modified, the validation chain breaks, signaling the asset's history has been altered.

03Why It Is Implemented

Platforms like Meta (Instagram, Facebook), Pinterest, and search engines leverage C2PA to automatically detect AI-generated content. When their servers parse a valid C2PA manifest showing an AI origin, they automatically append "Made with AI" labels or decrease search distribution visibility (shadowbanning).

Implementing tools like ScrubAI allows creators to dismantle this continuous tracking infrastructure. By breaking the marker stream and regenerating the image at a raw pixel level, we ensure absolute digital sovereignty, preventing third-party platforms from indexing and penalizing creative exports.