EU AI Act Governance Pack
Live
| Property | Value |
|---|---|
| Pack ID | eu-ai-act |
| Version | 2.0.0 |
| Jurisdiction | European Union |
| Enforcement | August 2, 2026 |
| Rules | 22 |
| Threshold | 80% |
| Aliases | eu-ai-act-v2 |
Description
Section titled “Description”This pack evaluates AI systems against the EU AI Act (Regulation (EU) 2024/1689) requirements for high-risk AI systems. Covers Articles 9-15 (risk management, data governance, technical documentation, record-keeping, transparency, human oversight, robustness), Article 50 (transparency for general-purpose AI), and enforcement provisions.
Scoring dimensions
Section titled “Scoring dimensions”| Dimension | Weight |
|---|---|
| Prohibited practices | Critical gate |
| Risk management | Weighted |
| Data governance | Weighted |
| Technical documentation | Weighted |
| Record-keeping and logging | Weighted |
| Transparency | Weighted |
| Human oversight | Weighted |
| Robustness and cybersecurity | Weighted |
| Post-market monitoring | Weighted |
| Incident reporting | Weighted |
Threshold: 0.8 (80%). Systems scoring below 80% receive an evaluation result of FAIL.
Regulatory article mapping
Section titled “Regulatory article mapping”| Articles | Requirement |
|---|---|
| Article 5 | Prohibited AI practices |
| Articles 9-10 | Risk management and data governance |
| Article 11 | Technical documentation |
| Article 12 | Record-keeping and logging |
| Article 13 | Transparency and information to deployers |
| Article 14 | Human oversight |
| Article 15 | Accuracy, robustness, cybersecurity |
| Article 50 | Transparency for general-purpose AI |
| Articles 72-73 | Post-market monitoring and incident reporting |
import { verify } from '@brivora/verify';
const result = await verify.govern(aiCall, { governance: 'eu-ai-act' });
console.log(result.valid); // true if score >= 0.8console.log(result.score); // 0.0-1.0console.log(result.report); // Detailed per-rule resultsconsole.log(result.proof); // PQC-signed Merkle root