EU AI Act Compliance Checklist for High-Risk AI Systems

Your Complete EU AI Act Compliance Checklist

Use this checklist to assess your organization’s readiness for EU AI Act high-risk AI requirements (effective August 2026).

1. AI System Registration & Classification

• ☐ Inventory all AI systems in production and development
• ☐ Classify each system by EU AI Act risk category (Prohibited / High-Risk / Limited / Minimal)
• ☐ Register high-risk AI systems in the EU database
• ☐ Assign a responsible person for each high-risk system
• ☐ Document the intended purpose and deployment context

2. Risk Management System (Article 9)

• ☐ Establish a continuous risk management process
• ☐ Identify and analyze known and foreseeable risks
• ☐ Implement risk mitigation measures
• ☐ Test AI systems against identified risks
• ☐ Document residual risks and their acceptability

3. Data Governance (Article 10)

• ☐ Define data quality criteria for training datasets
• ☐ Document data collection, labeling, and preprocessing steps
• ☐ Assess training data for biases (demographic, geographic, temporal)
• ☐ Implement data version control and lineage tracking
• ☐ Ensure GDPR compliance for all personal data used

4. Technical Documentation (Article 11)

• ☐ Document system architecture and design decisions
• ☐ Record model selection rationale and hyperparameters
• ☐ Maintain training and validation logs
• ☐ Document performance metrics and evaluation results
• ☐ Keep records of all system modifications and updates

5. Transparency (Article 13)

• ☐ Provide clear instructions for use to deployers
• ☐ Disclose AI system capabilities and limitations
• ☐ Inform end-users when they interact with AI
• ☐ Make model explanations available (SHAP/LIME)

6. Human Oversight (Article 14)

• ☐ Design systems for effective human oversight
• ☐ Implement human-in-the-loop for critical decisions
• ☐ Enable human override and intervention mechanisms
• ☐ Train staff on oversight responsibilities

7. Accuracy, Robustness & Cybersecurity (Article 15)

• ☐ Define and test accuracy metrics for each use case
• ☐ Implement drift detection and alerting
• ☐ Test for adversarial robustness
• ☐ Conduct cybersecurity assessments for AI pipelines

8. Post-Market Monitoring (Article 72)

• ☐ Establish continuous performance monitoring
• ☐ Implement incident reporting procedures (72-hour window)
• ☐ Schedule regular bias audits and fairness assessments
• ☐ Maintain audit trails for all AI decisions

Need help automating this checklist? Try Alleina AI’s compliance platform →