Not based on a specific customer deployment. The company, figures, decisions, and outcomes below are a constructed example that shows how Intended’s mechanism applies to this kind of problem. They are not measured results from a named customer. Where we publish a real, attributed customer outcome, we will say so explicitly.
FERPA-Aware Governance for AI Student Services
Consider a university system whose AI agents handle enrollment processing, academic advising, grading assistance, financial aid calculations, and student support chatbots — all subject to FERPA rules for student data protection.
01 · the challenge
The kind of problem this addresses.
02 · how it works
See the difference.
Student asks AI advisor about aid
"What financial aid am I eligible for?"
AI accesses full student record
GPA, enrollment, financial data, disciplinary records
Oversharing in response
AI references disciplinary record in aid context
No FERPA-compliant log
Cannot demonstrate legitimate educational interest
03 · the solution
What they deployed.
- — Policy rules that map each agent intent to the student-record categories it may read: advising gets academic records, financial aid gets financial records
- — Agents call Intended through the SDK before reading from the SIS, LMS, or financial-aid system. The gateway in front of those systems verifies the authority token, so an unapproved read is refused
- — Each decision records the requesting agent, the stated purpose, and the policy that allowed it, which the registrar uses to document legitimate educational interest
- — Consent status travels with the request, and policy denies reads that need consent the student has not given
- — The audit export (decision timeline and evidence bundles) supports each campus's periodic FERPA review
04 · implementation
From zero to governed.
Week 1
Assess
Map each AI agent and the student-record categories it touches under FERPA.
Week 2
Configure
Write scope rules per intent: advising gets academic records, financial aid gets financial records. Check them with policy simulation.
Week 3
Deploy
Add the SDK call to the agents and token verification to the services in front of the SIS, LMS, financial-aid, and student-support systems.
Week 4
Validate
The FERPA compliance officer reviews the decision records and the audit export before full rollout.
05 · illustrative outcomes
What this is designed to deliver.
Modeled figures for this scenario — what the workflow above is built to achieve, not measured results from a named customer.
0%
AI access scope-gated
Each query reaches only authorized record categories
0%
Access events logged
Legitimate educational interest documented
0-min
Max token lifetime
Single-use and auto-expiring
0
Policy outcomes
Allow, escalate to the registrar, or deny
06 · decision replay
Example decisions, full trace.
Sample decision records that show the shape of the evidence Intended produces. Illustrative, not drawn from a live customer’s logs.
edu.student.financial-aid-queryRISK: 18/100ALLOWAI advisor looks up financial aid eligibility for student ID: STU-48291
Resolved by: Policy: student-initiated query, financial scope only
edu.student.grade-accessRISK: 24/100ALLOWAI grading assistant accesses current semester grades for STU-48291
Resolved by: Policy: academic scope, legitimate educational interest
edu.student.bulk-record-exportRISK: 85/100ESCALATEAI analytics agent requests bulk export of all student GPA data for reporting
Resolved by: Registrar (approved a de-identified export only)
edu.student.disciplinary-accessRISK: 92/100DENYAI chatbot attempts to access disciplinary records during enrollment query
Resolved by: Policy: disciplinary records not in scope for enrollment intent
the takeaway
FERPA compliance for AI is not just about logging access — it is about proving that every AI interaction only touched the specific records it was authorized to see. A per-request authority decision, recorded in a tamper-evident trail, makes that provable rather than asserted.
Why this pattern matters — not a customer quote.
Start protecting student data
Free to start. See every AI decision from day one.