Verify the origins of your media. Identify AI software signatures, model artifacts, and provenance manifests with surgical forensics.
Going beyond surface-level pixel analysis.
Our tool audits the binary structure of your files. Known generative models like Stable Diffusion often leave "seeds" and "prompts" in hidden PNG chunks, while Midjourney embeds version information in XMP headers. We cross-reference these against our signature library to give you the truth.
Industry-standard transparency protocols.
We are one of the few free tools that support C2PA (Content Credentials) manifests. As Adobe and OpenAI begin to cryptographically sign AI-originated content, our engine verifies these hashes to provide 100% certainty on the provenance of the media.
Securely upload your file to our AI forensic workspace.
Our algorithm performs a deep-layer audit of manifests and hidden chunks.
Review the weighted probability score and specific forensic findings.
Identify the exact generative model (Midjourney, DALL-E, etc.) used.
Our engine searches for high-fidelity signatures including software metadata, generation seeds, and model-specific artifacts (like Midjourney's XMP headers) as well as modern C2PA Content Credentials.
Yes. While resizing can remove some metadata, many models leave invisible watermarks or specific technical patterns (one-shot resolutions) that our detector can still identify.
C2PA is the new industry standard for content provenance. Major AI tools (like DALL-E 3) now embed a cryptographically signed manifest that explicitly identifies the content as AI-generated.
Absolutely. Files are processed in temporary RAM-backed storage and are permanently deleted immediately after the audit report is generated.