China's Five Ministries Launch "AI + Education" Plan: Full Rollout in K-12 by 2026
On April 10, 2026, five Chinese government ministries — Education, NDRC, MIIT, MOST, and National Data Administration — jointly issued the "AI + Education Action Plan."
This is China's first education policy that positions AI as "infrastructure." Not a pilot program — a full-scale national rollout.
What the Policy Says
The core target: by 2027, 100% AI education coverage in K-12 schools, and full AI literacy course coverage in higher education.
Three key points:
One, AI in the classroom isn't about teaching coding. The policy emphasizes using AI to assist teaching, not teaching students AI technology. For teachers: AI grading, generating tiered exercises, analyzing student knowledge gaps. For students: AI personalized learning paths, adaptive quiz banks, intelligent speaking practice.
Two, infrastructure first. By end of 2026, all K-12 schools nationwide should be connected to the "AI Education Cloud Platform." Build the network first, applications later. Each region must establish computing resource pools, prioritizing AI services for education.
Three, data security and privacy are red lines. Student data stays within the campus network. AI models deploy locally. Commercial AI products entering schools require regulatory review. This is why the policy stresses "controllable" — not just any AI can enter schools.
Industry Impact
Tutoring companies are the first to feel the shift.
The policy encourages "school-external collaboration" but sets a bar: AI education products entering schools must pass Ministry of Education filing, with local data deployment. This effectively eliminates small-scale AI education tools. Big tech companies with technical and compliance capabilities will win.
Hardware manufacturers benefit. Demand for classroom AI terminals (smart screens, voice interaction devices, AI cameras) will surge. Seewo, Hitevision, and other education hardware players are already racing for contracts.
Public school teachers' workflows will change. AI essay grading, lesson plan generation, learning analytics — these aren't concepts, they're real scenarios already proven in Shanghai and Shenzhen pilots. The 2026 push into tier-2/3/4 cities will make teacher training the biggest execution bottleneck.
But there's an unspoken concern: will AI-assisted teaching make teachers lazy? If teachers get used to AI generating exercises, AI grading, AI writing comments — will the "human touch" of teaching be lost? The policy document doesn't address this, but it will inevitably surface in practice.
Where the Opportunities Are
If you're in edtech, several tracks are worth watching:
- AI lesson plan / courseware generation: A teacher's biggest time sink. A good tool here has enormous demand
- AI learning analytics: Using AI to analyze student homework and exam weak points, generating personalized remediation plans
- AI speech assessment: AI practice and evaluation for English speaking tests and Mandarin proficiency exams
- Campus AI terminals: Hardware + software integration for smart classrooms
But note: entering public schools means long decision cycles, strict compliance requirements, and potentially thin margins. Payment cycles start at one year. Not everyone can wait that long.
Consumer-facing AI education products (AI tutors, AI homework assistants) don't need to wait for policy — parents pay directly. But competition is already fierce.