Quality standards
Not every task is easy. But every approved task is tracked.
Every project has quality rules. Accuracy matters more than speed. Here is exactly what reviewers check and why rejections happen.
What review usually checks
- Accuracy against the project instructions
- Completeness (nothing required was skipped)
- Duplicate risk (the same work submitted twice)
- Project-specific rules, like timestamp tolerance
- Privacy rules for recording projects
Common rejection reasons
- Tags missing or outside the allowed label list
- Wrong or drifting timestamps
- Unclear video segment tagged anyway instead of flagged
- Privacy rule not followed on a recording
- Duplicate upload
- Incomplete checklist
- Low accuracy score
Rejections are feedback, not punishment
If your submission does not match the instructions, reviewers may ask for corrections or reject it, with feedback wherever possible. If you believe a review is wrong, you can raise a rejection dispute and a different reviewer re-checks the work. Consistent quality is what unlocks better projects.
How quality unlocks levelsStart with your first verified AI project.
Create your profile, complete secure verification, and begin training. Project availability may vary by city and demand.
Join the early access list