Make AI content show its work.
Mark every AI output with standard, machine-readable provenance. Prove what your AI made, when, and with what. Built for the EU AI Act marking obligation: August 2, 2026 for new systems, December 2, 2026 for systems already on the market.
How it works
Connect your pipeline
One API call where your content is generated. An afternoon, not a quarter.
Every output carries its mark
A C2PA manifest travels invisibly inside the content and survives copy-paste.
See your coverage
Your dashboard shows what share of outputs are marked, ready to show an auditor.

Where this fits
Security and data handling
Standard manifests, no lock-in
Marks are C2PA manifests. Regulators, platforms, and standards-based verifiers can read them without Encypher in the loop.
Audit trail for your evidence file
Every marked output is logged; coverage reporting and a compliance dossier are built into the dashboard.
For engineers
curl -X POST https://api.encypher.com/api/v1/sign \
-H "Authorization: Bearer $ENCYPHER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"text": "Your AI-generated output here.", "metadata": {"model": "gpt-4o"}}'- Three API calls: sign outputs, scan inputs, analytics
- C2PA 2.x manifests with hard binding plus text soft binding
- Works for text, images, audio, video, and PDF
- 500 AI output documents per month free
Ready to mark your AI outputs?
Start free with 500 AI output documents per month. New generative systems must mark outputs from August 2, 2026; systems already on the market have until December 2, 2026.