Observed arrival · 2026-11-22
Qproov wants the work, not just the answer
Qproov describes an assessment platform that lets people use AI on practical tasks and records how they define, test, and complete the work.
- For
- Hiring teams and practical-skills candidates
- Worth noticing
- The page says keyboard contents are not stored; it separately labels identity checks from L0 to L2 and does not claim authorship proof.
Field notes
The proposed record covers plans, prompts, AI responses, accept-or-reject decisions, revisions, and execution, in sequence. Qproov says those events are sealed in a hash chain, while explicitly excluding keystroke contents and distinguishing identity checks from authorship proof. Its example pass shows three scored dimensions with ranges and reviewer count; the page says pilot scores are still uncalibrated and should not trigger automatic rejection.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue