RFC-028: EU AI Act Alignment — Regulatory Governance Framework — 7. General-Purpose AI Models (Art. 51-56)

AIGP Specification › RFC-028: EU AI Act Alignment — Regulatory Governance Framework › 7. General-Purpose AI Models (Art. 51-56)

← 6. Transparency Obligations (Art. 50) · Section index · 8. Post-Market Monitoring (Art. 72-73) →

7. General-Purpose AI Models (Art. 51-56)

7.1 GPAI Classification

{
  "gpai_context": {
    "is_gpai": true,
    "systemic_risk": false,
    "cumulative_compute_flops": 9.5e24,
    "systemic_risk_threshold": 1e25,
    "code_of_practice_adherent": true
  }
}

7.2 GPAI with Systemic Risk (>10²⁵ FLOPs)

Additional obligations:

• Model evaluation (adversarial testing, red-teaming)
• Serious incident reporting
• Cybersecurity protections
• Energy consumption reporting

---

---

← 6. Transparency Obligations (Art. 50) · Section index · 8. Post-Market Monitoring (Art. 72-73) →