How we earn the trust we ask of vendors
SafeGradeAI evaluates AI tools used by children. We hold ourselves to the same standards we publish — and document them here in plain language.
Encryption in transit & at rest
All traffic is served over TLS. Data is stored on managed Postgres with encryption at rest. Backups are encrypted.
Least-privilege access
Production data access is limited to a small set of admins, audited, and protected by single-sign-on and 2FA.
Row-level security
Every user-facing table uses Postgres row-level security so users only ever see their own records.
No selling of personal data
We do not sell visitor or member data. We do not run third-party ad networks on logged-in experiences.
COPPA-aware design
SafeGradeAI is not directed to children under 13. Our member accounts are for adults (parents, educators, administrators).
FERPA-aware reviews
When evaluating school-facing tools, we surface each vendor's posture on FERPA and major state student-privacy laws.
Reputable subprocessors
We use established providers for hosting, database, payments (Stripe), email, and analytics. Some are documented below.
Independent of vendors
We do not let vendors edit their own scores. Sponsored evaluations are clearly labeled and follow the published rubric.
Subprocessors
Third-party services we use to operate SafeGradeAI.
| Provider | Purpose | Data |
|---|---|---|
| Supabase | Database, authentication, file storage | Account & application data |
| Cloudflare | Hosting & edge runtime | Request metadata |
| Stripe | Payments & subscriptions | Billing data |
What we collect
- Account data: email, name, account type (parent / school / company), organization name.
- Usage data: aggregate analytics on which evaluations and pages are visited. We never sell this data.
- Submissions: tool review requests, sales leads, contact messages, and tool submissions you choose to send us.
- Payment data: handled by Stripe. We never see or store full card numbers.
Reporting a security concern
If you believe you have found a security issue, please email support@safegradeai.com. We respond to legitimate reports within two business days.