AI Watermarks
What Are AI Watermarks?
A primer on visible labels, invisible statistical watermarks, cryptographic provenance, and why a missing signal does not prove content is human-written.
Published
Read guideGuides
Explainers with a usable vocabulary
Guides on this site teach the difference between inspectable artefacts and claims that cannot currently be checked. Categories exist for later expansion; we will not mass-produce near-duplicate keyword pages.
Regulation & standards
Planned categories
- AI Watermarks
- C2PA
- SynthID
- Metadata
- Content Credentials — planned
- Providers — planned
C2PA
What Is C2PA and Content Credentials?
An explainer of the C2PA standard and Content Credentials: signed provenance attached to files, how it differs from invisible watermarks, and what a missing manifest does and does not mean.
Updated
Read guideSynthID
What Is a SynthID Watermark? Images, Text, Audio & Video
SynthID is Google DeepMind’s invisible watermarking technology for AI-generated images, text, audio, and video. Detection is provider-specific — AI Watermark Center does not detect SynthID.
Last verified
Read guideMetadata
Hidden & Invisible Unicode Characters in Text
What hidden and invisible Unicode characters are, why they appear in copied text, and what listing them can — and cannot — prove about AI. Use the text scanner when you need to check a string.
Updated
Read guide
Related pages
- EU AI Act & AI-generated contentArticle 50 transparency hub — not legal advice.
- C2PA vs SynthIDMetadata provenance versus invisible watermarks.
- AI watermark detector (what can be checked)No universal detector — route to real tools.
- Tools these guides supportInspectors that operationalise the vocabulary above.
- Provider profilesSourced provider research hubs.
- MethodologyThe evidence rules behind factual pages.