FERPA-Aware by Design
The safest way to handle student records in search is to not need them. Public site search runs on the pages you've already chosen to publish, so a transcript or a grade never has to enter the index for search to work. Where you do connect an authenticated source, access controls are respected at query time, so a visitor only ever sees what they were already allowed to see. Your data stays in the US, and none of it trains an AI model.
How it works
Index only what's public
The crawler reads the pages you publish. A student's record isn't in scope because public search never needed it. The sensitive data simply stays out of the index.
Honor permissions where they apply
When you connect an authenticated or restricted source, Keyspider checks access at query time. Two people searching the same term see only the results each is cleared for.
Keep the data in its lane
Everything stays in US data centers. Queries are anonymized, nothing is sold or shared, and your content is never fed into model training.
Use cases
University public sites
A campus search spans admissions, academics, and services without touching the student information system, so FERPA exposure on the public site is designed out from the start.
Authenticated student portals
Inside a logged-in portal, access-aware results mean a student finds their own forms and holds, and never stumbles onto another student's.
Procurement and privacy review
A privacy office can confirm that public search needs no protected records, that data stays in the US, and that nothing trains an AI, which answers most of the questionnaire up front.
Ready to give your users better answers?
AI Search, AI Assistant, and Workplace Search. Deployed in days, not months. See it live on your own content.
No credit card required · Live in 2 weeks · Cancel anytime