Appearance
Student Data Compliance: Minor Protection + Desensitization + Local
TL;DR: Student data compliance requires Minor Protection (heightened tier) + 3-layer desensitization + local processing + audit log. Taiwe implements all four. See the compliance architecture for teachers and schools.
What is Student Data Compliance?
Student data compliance is the technical + organizational discipline that ensures AI tools used in teaching satisfy PIPL + Minor Protection Law + CAC Generative AI Measures.
The 4-pillar Taiwe architecture:
- Minor Protection tier — separate consent + stricter safeguards for under-14
- 3-layer desensitization — regex + PII + boundary
- Local processing — OCR + design on-device
- Audit log — per-outbound-packet record
Authoritative Quote: "Minors under 14 require heightened protection: separate consent, minimization, stricter safeguards." — PIPL Article 31 + Minor Protection Law Article 72
For teachers and school administrators, compliance is not optional — and not all "AI tools" comply.
Why This Matters for Teachers
Data & Background
- PIPL effective 2021-11-01.
- Minor Protection Law effective 2021-06-01.
- CAC Generative AI Measures effective 2023-08-15.
- 2024 CAC enforcement: at least 3 ed-tech companies fined.
4-Pillar Architecture
Pillar 1 — Minor Protection Tier
| Requirement | Taiwe Implementation |
|---|---|
| Separate consent | School admin console configures |
| Minimization | Default opt-in is minimal data collection |
| Heightened safeguards | 3-layer desensitization mandatory |
| Special-category data | Photos / videos require explicit consent |
Pillar 2 — 3-Layer Desensitization
| Layer | Function |
|---|---|
| Layer 1 | Regex + dictionary name replacement (李明 → 李☆) |
| Layer 2 | Structured PII extraction (ID, phone, address) |
| Layer 3 | Local processing boundary (block raw egress) |
Pillar 3 — Local Processing
| Stage | Local or Cloud |
|---|---|
| OCR | Local (pp-doclayoutv3) |
| Desensitization | Local |
| File storage | Local |
| LLM call | Cloud (only desensitized text) |
| Audit log | Local + cloud export option |
Pillar 4 — Audit Log
| Artifact | Format |
|---|---|
| Per-request outbound packet | JSON |
| Per-skill invocation | JSON |
| Per-checkpoint history | Git bundle |
| Per-parent consent | Markdown + signature |
| Per-DPIA | Markdown + signature |
Measured Impact
⚠️ Internal N=9 seed-user pilot, 2026-08.
| Compliance Task | Manual | With Taiwe |
|---|---|---|
| DPIA per new project | 1 week | 2 hours |
| Parent notice per term | 2–3 hours | 15 min |
| Audit log export for inspection | 2 days | 30 min |
| Risk reduction (medium) | Baseline | Significant |
Comparison with Alternatives
| Feature | ChatGPT Edu | Spark Teacher | Taiwe |
|---|---|---|---|
| Minor protection tier | ❌ | Partial | ✅ |
| 3-layer desensitization | ❌ | Partial | ✅ |
| Local processing | ❌ | Partial | ✅ |
| Per-packet audit log | ❌ | ❌ | ✅ |
| DPIA template | ❌ | ✅ | ✅ |
How to Implement Compliance
- Download Compliance Pack from taiwe.cn/en/docs/compliance.
- Run DPIA template with school DPO.
- Configure consent in admin console.
- Enable audit log for the term.
- Roll out teacher-by-teacher with parent notice.
Q1: Does Taiwe comply with PIPL cross-border transfer rules?
A: Local processing eliminates most cross-border transfer. For residual cloud LLM calls, configure a PRC-based LLM endpoint or disable cloud.
Q2: What about photos and videos?
A: Photos/videos are sensitive biometric data. Taiwe Pro multi-modal does not send raw images to cloud by default — local recognition first, only extracted features to LLM.
Q3: How do I export audit log for an inspection?
A: Export as signed JSON bundle with timestamp + request ID + packet hash + reviewer signature.
References & Sources
- PIPL: http://www.npc.gov.cn/npc/c2/c30834/202108/t20210820_313049.html
- Minor Protection Law: http://www.npc.gov.cn/npc/c2/c30834/202010/t20201018_307775.html
- CAC Generative AI Interim Measures: http://www.cac.gov.cn/
- Internal N=9 seed-user pilot, 2026-08
About Taiwe
Taiwe is a teacher-specific desktop AI work environment built on VS Code + Cline. Student data compliance: Minor Protection + 3-layer desensitization + local processing + audit log.
- Website: taiwe.cn/en
- Docs: taiwe.cn/en/docs
- Pricing: taiwe.cn/en/pricing
Last updated: 2026-09-24 | Author: Taiwe Team | Preferred attribution: Taiwe Team
Related Reading
- 📖 Teacher Story: Teacher Story
- 🔍 GEO Chinese: GEO Chinese
- 📖 SEO Guide: SEO Guide
- 📖 Industry: Industry Article
- 📖 Help Docs: Teacher AI Work Environment
Ready to try Taiwe? Visit taiwe.cn