Claude model check
Claude Sonnet watermark detector
Check Claude Sonnet-related watermark, provenance, and AI-writing risk signals while tracking public support limits.
Run a signal check
No detector model loads until you press Check.
Current support
Future Claude models, including Sonnet-class outputs when enabled, are expected to use Anthropic's text watermarking, but public verification is still pending.
What this checks
- Anthropic says text watermarking applies to words Claude chooses, including generated translations.
- Model-specific pages preserve the claimed model in server and saved reports.
- Reports should separate Sonnet watermark availability from generic AI text risk.
What this does not prove
- Claude Sonnet outputs may not expose public watermark verification until Anthropic's detection API is available.
- Short passages, light proofreading, exact factual output, and code can all reduce watermark detectability.
- Heavy edits can weaken both watermark and classifier evidence.
FAQ
- Can this prove whether content was written by AI?
No. AI Detect reports signals such as provenance metadata, watermark support status, and statistical text risk. It does not claim legal or authorship proof.
- Does text leave my browser during a local check?
Browser-default checks run local text features and provenance checks after you press Check. Server enhanced checks are separate and will be labelled before use.
- Why can a result be unknown?
Many providers do not expose public watermark detectors, and metadata can be absent or stripped. Unknown is a valid result when no reliable public signal exists.
Workflow upgrade
Need repeatable reviews?
Move from one-off checks to saved reports, batch review, and API-ready signal records when the workflow needs it.