Privacy score: 92/100. Transparency score: 86/100. Overall 91/100 (KidSafe Approved), reviewed 2026-07-15.
What the vendor documentation says
SchoolAI publishes a Privacy Policy and Trust Center documenting a Data Processing Addendum for schools, SOC 2 Type II, and student-privacy commitments. Student prompts and responses are not used to train foundation models; SchoolAI acts as a service provider / school official under district contracts. COPPA: designed for K-12 with school-consent provisioning for under-13 students, in line with the FTC's school-authorized-consent framework. FERPA: SchoolAI signs a DPA with districts and operates as a school official with legitimate educational interest. State privacy laws: SchoolAI is listed as signatory or compliant with the Student Data Privacy Consortium (SDPC) National Data Privacy Agreement (NDPA) in many participating districts. GDPR: SCCs are available for EU districts. EU AI Act: SchoolAI publishes safety and use documentation consistent with GPAI transparency expectations.
Controls and safeguards on the data path
Safeguards are aligned with a K-12 supervised model: teacher-created assignment scope, real-time monitoring dashboards, per-student session records, automatic flagging with severity levels, and admin controls at the district/school level. Sign-in is via SSO (Google/Microsoft/Clever) so students use school-managed accounts. Session records are retained for teacher review and district audit. Weaknesses: because scope and safety depend on how teachers configure Spaces, a poorly configured Space can still let students steer far off task; SchoolAI mitigates this with default prompts and templates but district training matters. No per-student time limits; product breadth is expanding, so admins should re-review new features quarterly.
Open privacy concerns
- Safety depends on how well teachers configure and monitor Spaces — training matters.
- High-volume classes can generate more flags than a single teacher can triage in real time; district-level triage should back up the classroom.
- Rapid feature expansion (assistants, teacher tools, admin analytics) means new capabilities may launch with defaults worth reviewing.
- Underlying LLM responses can still hallucinate; students need media-literacy framing on top of the safety layer.
- No per-student time or session limits — pacing is the teacher's responsibility.
Questions to ask before you sign
- Is the DPA in place, and are we listed under the SDPC National Data Privacy Agreement?
- Is SSO enforced through Google, Microsoft, or Clever for all student accounts?
- What is the escalation path from a teacher flag to counselor or administrator review?
- What baseline teacher training on Space configuration and flag review is required before rollout?
- How often do we re-review newly launched features against district AI policy?
District posture we recommend
Policy recommendation for K-12 administrators: approve SchoolAI with (1) a signed Data Processing Addendum (SchoolAI's standard or your state's NDPA) covering FERPA, COPPA, and applicable state student-privacy laws, (2) SSO enforced through Google, Microsoft, or Clever so all student accounts are school-managed, (3) an escalation path from teacher-level flags to counselor/admin review for high-severity categories (self-harm, abuse disclosures), (4) a written expectation that teachers monitor the live dashboard during Space use, (5) baseline teacher training on Space configuration and flag review, and (6) quarterly review of new SchoolAI features against district AI policy. Block personal ChatGPT/Gemini/Copilot access on student devices to keep AI use inside the supervised, DPA-covered environment.
What this article does not claim
Findings labelled as verified come from SafeGradeAI's published evaluation of this tool and the vendor's own public documentation as of the review date above. Anything framed as a recommendation is our professional guidance, not a statement of fact about the product. Vendor policies, model versions, and default settings change frequently — re-check the vendor's current documentation before making a district-level decision, and treat any AI output as a draft to verify.
This guide is built on the SafeGradeAI evaluation of SchoolAI. Read the full SchoolAI evaluation and rubric scores.
Frequently asked questions
Can teachers really see every message?+
Yes. Real-time teacher dashboards show every student conversation, and teachers can enter a conversation, redirect it, or end the session.
What gets flagged?+
The system flags concerning content across categories including self-harm, violence, bullying, sexual content, and disclosures of abuse, and surfaces flags to the teacher and, where configured, to district administrators.
Is student data used to train AI models?+
No. The evaluation records that student prompts and responses are not used to train foundation models, and SchoolAI operates as a service provider / school official under district contracts.
What is the main weakness?+
Safety depends on configuration and attention. A poorly scoped Space can drift off task, and a high-volume class can generate more flags than one teacher can triage live — district-level triage should back up the classroom.
Recommended next steps
For schools
Govern AI tools district-wide with School Pro
Unlimited full evaluations plus the K-12 workspace — tool registry, approval workflow, monitoring, and a downloadable audit log.
For parents
Start with the Family AI Safety Starter Pack
A free, practical pack you can use alongside SchoolAI — conversation starters, age guidance, and a family AI agreement.
Want the deep dive? Get the Complete AI Safety Blueprint — $29