About AI Inspector Lab

Inspect. Detect. Verify. - in that order, and with the evidence on the table.

Why we exist

Generative AI made it trivial to produce convincing text, images and code. The tools that claim to detect it often return a single, confident percentage with no explanation. Decisions about people - grades, hiring, publication - are then made on numbers nobody can check.

AI Inspector Lab takes a different approach. We treat detection as an investigation: gather independent evidence, show where it came from, state how reliable it is, and be explicit about what cannot be known.

Our principles

• No fabricated scores. If no reliable detection model is configured, the likelihood is reported as unavailable.
• Evidence stays separate. Model estimates, Content Credentials, metadata and heuristics are never blended into one number.
• Probabilistic, not proof. Every report carries its limitations, and our product copy never claims certainty.
• Portable and private. Uploaded files and pasted text are private to your account and deleted on request (free scans of public websites and repositories appear in the public Explorer; Pro and Team keep them private); the platform runs on open standards and can move between infrastructure providers.

Get in touch

Questions, partnerships or research collaborations: contact us.