Remove C2PA Content Credentials with a structural strip
AI Stripped detects C2PA/JUMBF containers and strips provenance-bearing segments (JPEG/PNG walks, re-encode, lossless path when supported). This is not a cryptographic forgery toolkit and not a guarantee against future re-attachment.
Results are best effort. Residual risk is retained, originals are preserved, and nothing is marketed as undetectable.
What this pass can do
- Structural detection of C2PA / JUMBF markers
- Lossless provenance strip for supported JPEG/PNG when mode=metadata
- Sharp re-encode strip for other containers
- Honest report language, with residual risk retained
Limits you should know
- Processing is best-effort. Embedded watermarks are designed to survive edits.
- We never claim 100% undetectable results or a guaranteed public-detector pass.
- Absence of a known signal does not prove content is human-made.
- You remain responsible for rights, provider terms, and disclosure laws.
FAQ
- Is this the same as c2pa-rs validate?
- No. We do structural detect/strip, not full signature lifecycle management. Validation of third-party claims stays external.
Related guides & tools
- How to remove C2PA / Content Credentials: the full structural-strip walkthrough
- ChatGPT / DALL·E C2PA strip: OpenAI-class exports emphasize C2PA
- AI provenance signals explained: why C2PA is metadata, not a watermark
- AI watermark remover: the full multi-layer pipeline