OpenAI model family checks
GPT watermark detector
Check GPT-related AI-writing, provenance, and watermark support signals with clear uncertainty.
Run a signal check
No detector model loads until you press Check.
Current support
Public GPT watermark detection support varies by generation method. AI Detect reports available signals and limitations.
What this checks
- GPT-related pages should cover ChatGPT, GPT-4o, GPT-5, and o-series variants through the model registry.
- Legacy RoBERTa detectors are labelled as statistical baselines, not GPT proof.
- Provenance metadata is checked separately from text likelihood.
What this does not prove
- OpenAI-specific watermark detection may not be public for arbitrary pasted text.
- Legacy detectors can be outdated and must be labelled carefully.
- False positives are possible, especially for formal or non-native writing.
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.