Identify what would leave the device
Before server OCR, a Kenyan school should review the actual documents and the service's current file-handling information. Scanned files can contain learner names, staff details or private annotations that do not belong in an ordinary exam-cleanup upload. Use the minimum authorised material for a trial.
Distinguish local browser work from temporary server processing. Read the site's current security information and ask support about any unresolved retention or access question. Do not infer a specific deletion time from an old screenshot or a general marketing statement.
Choose the appropriate handling route
Where possible, use a non-sensitive representative sample to establish the workflow. If sensitive data is involved, follow the school's authorised privacy and data-handling process before uploading. Cosmetic removal of names is not proof that a file is safely anonymised.
Keep a record of what was approved for processing and how the output will be stored. This guide does not determine legal compliance or certify the service's current infrastructure. The institution must verify the facts and obtain appropriate advice where required.
Practical checks
- Review visible and hidden information
- Identify server transfer
- Read current handling details
- Minimise the trial sample
- Document the school's authorisation
Questions and answers
Can I assume a fixed automatic-deletion period?
No. Confirm the currently configured and documented practice.
Does erasing visible names guarantee anonymisation?
No. Review other identifying and recoverable information appropriately.