Dubai's private-school regulator KHDA and the UAE's national PDPL framework now require age-staged, teacher-supervised AI access, mandatory Data Protection Impact Assessments before deployment, and a ban on vendors retraining models on student data — though KHDA itself admits no
Details
Maturity
Scaling
Promoter
Knowledge and Human Development Authority (KHDA), Dubai
Period
2025–ongoing
Keywords
AI policy, data protection, age-staged access, private-school regulation, PDPL compliance
Context
Dubai's Knowledge and Human Development Authority (KHDA), which regulates the emirate's large private-school sector, has built generative-AI governance around a staged, age-based access model rather than a blanket ban or blanket permission. Early years and lower-primary students get no independent access to open generative-AI tools — usage is teacher-led. Upper-primary and lower-secondary students are introduced gradually in supervised, group-based activities teaching ethical prompting, bias recognition and fact-checking. Only secondary and older students move to supervised individual use, and AI use is banned outright during formal assessments across all age groups.
Activities
Schools must document what personal student data they collect, where it is stored and who has third-party access; no personally identifiable or sensitive health/behavioural data may be entered into public models; vendors must be contractually barred from using student inputs to retrain their models; and schools are expected to run Data Protection Impact Assessments before deploying new AI services, under the UAE's federal Personal Data Protection Law (PDPL). This sits alongside a separate national policy: in May 2025 the UAE Ministry of Education made AI instruction compulsory in public schools from kindergarten through secondary school.
Conclusions
The framework is honestly incomplete by KHDA's own account: a school principal told reporters there is presently no approved or centrally issued list of AI platforms from KHDA for school use, leaving individual schools to conduct their own due diligence — a real gap between the stated principles and centralised enforcement. As KHDA's Rumaisah Wajid put it, 'The challenge is no longer access to tools, but ensuring their quality, safety, and relevance.'
Implementation
Indicative cost
Medium (€50k–€500k) — No public cost figures disclosed; cost_band set to medium reflecting the compliance burden (DPIAs, staff training, contractual review) placed on schools network-wide.
Time to results
Long (> 3 years) — Ongoing regulatory framework since 2025, alongside a parallel national K-12 AI-instruction mandate from May 2025; timeline_band set to long.
Staffing & skills
School-level compliance staff conducting due diligence on AI tools, KHDA regulators setting and monitoring the framework
Conditions for success
Age-staged access tiers (teacher-led → supervised group use → supervised individual use)
Mandatory Data Protection Impact Assessments before deployment
Contractual bans on vendors retraining models on student data
AI banned outright during formal assessments
Common failure modes
No centrally approved or issued list of AI platforms yet, leaving schools to their own due diligence
No specific provision found addressing equity across fee levels between well-resourced and smaller private schools
Where it fits
Governance type
emirate-level private-school regulator (KHDA) plus UAE federal PDPL law
Scale
Dubai's large private-school sector
Data sources
Where this practice's information was retrieved from, and when.
Finland's education ministry and Opetushallitus published national AI recommendations (31 March 2025) after consulting 172 organisations and youth 'Digiraati' councils, …
Portugal's National Education Council published Recommendation 4/2026 in May 2026, mandating AI literacy from primary school, an independent education AI …