AI Ethics & Governance Checklist

UAE Government AI Projects — Pre-Deployment Sign-Off

Project Name: _______________
Entity: _______________
AI Champion: _______________
Review Date: _______________

Section 1: Data & Privacy

• Personal data used in training/inference is consented or appropriately anonymised
• Data storage complies with UAE Data Protection Law (Federal Decree-Law No. 45 of 2021)
• No cross-border data transfer without approval (for sensitive government data)
• Data retention policy defined and enforced
• Data lineage documented (where did the training data come from?)

Section 2: Fairness & Bias

• Model has been tested across demographic groups (gender, nationality, age) where applicable
• Bias testing results documented and reviewed by a second team
• Any identified bias has been mitigated or explicitly accepted with justification
• Monitoring plan in place to detect bias drift after deployment

Section 3: Transparency

• Citizens/users are informed they are interacting with an AI system
• AI-generated decisions or outputs are clearly labelled as such
• Explainability mechanisms in place for high-impact decisions
• Documentation available for regulators on request

Section 4: Human Oversight

• Human review process defined for high-stakes decisions
• Override mechanism exists (human can overrule AI output)
• Escalation path documented and tested
• Responsible AI owner assigned and accountable

Section 5: Security

• Penetration testing completed on AI endpoints
• Prompt injection protection in place (for LLM-based systems)
• Model output logging enabled for audit
• Access controls on model and training data reviewed

Section 6: UAE Alignment

• Aligned with UAE National AI Strategy 2031
• TDRA AI Guidelines reviewed
• Arabic language support verified (if public-facing)
• Entity leadership formally approved deployment

Approved by: _______________
Signature: _______________
Date: _______________