FAQ · AI content & watermarks

What is C2PA / Content Credentials?

Short answer

C2PA is a cryptographic provenance standard; its signed “Content Credentials” manifest is embedded in the file and tells platforms whether AI made it.

C2PA — the Coalition for Content Provenance and Authenticity — is a standard co-founded by Adobe, Microsoft, BBC, Sony and others that cryptographically signs a file with information about how it was made, including whether AI generated it. That signed manifest is called Content Credentials and is embedded as a JUMBF binary block inside PNG, JPEG and MP4 files. Tools like OpenAI’s Sora and DALL-E, and Adobe Firefly, attach it by default. Platforms such as Instagram and X can verify the signature in milliseconds and flag anything marked “made with AI.” It survives basic edits, so it has to be stripped deliberately.

Related on Calabi: Does Instagram detect AI images? · How do I remove C2PA metadata? · Does cropping remove a watermark? · How to post AI content on Instagram without getting shadow-banned · Sora — or just open the tool.

Strip it in one pass. 10 free cleans with a forensic proof card.
Try free →
Sources & references