What EU AI Act Article 51 - Classification of GPAI Models with Systemic Risk requires
EU AI Act Regulation 2024/1689 Article 51 establishes that a general-purpose AI model is classified as a GPAI model with systemic risk where it is trained using a total computing power of more than 10^25 floating point operations (FLOPs), that such models are presumed to have systemic risk and are subject to the enhanced obligations under Article 55, that the Commission may update the FLOPs threshold by delegated act after consulting the AI Board, that the AI Office may classify a GPAI model below the threshold as having systemic risk based on qualitative criteria including cross-sector impact, reach of the model, and the degree of autonomy and irreversibility of its outputs, and that providers may rebut the systemic risk presumption by demonstrating to the AI Office that their model does not pose systemic risk.
Pillar: AI Governance & Law · Authority: European Commission · Version: 1.0.0 · Last updated:
Primary source: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32024R1689
SHA-256 integrity: 3276e1e2e0f08ad779015c7f4fc83d582aee65b1dbcedf6a7711bba8aa4b59b5
Primary Citations — 5 traced to source
- {"title":"Regulation (EU) 2024/1689 - Article 51: Classification of GPAI Models with Systemic Risk","url":"https://eur-lex.europa.eu","relevance":"Primary EU AI Act provision establishing the 10^25 FLOPs quantitative threshold for systemic risk classification of GPAI models, the presumption of systemic risk for models exceeding the threshold, the Commission's power to update the threshold by delegated act, and the AI Office's power to classify sub-threshold models as having systemic risk on the basis of qualitative criteria including cross-sector impact, reach, autonomy, and irreversibility of outputs"}
- {"title":"Regulation (EU) 2024/1689 - Article 55: Obligations for Providers of GPAI Models with Systemic Risk","url":"https://eur-lex.europa.eu","relevance":"EU AI Act provisions establishing the enhanced obligations that apply to providers of GPAI models classified as having systemic risk under Article 51, including model evaluation, adversarial testing, serious incident reporting, cybersecurity measures, and energy consumption reporting obligations triggered by the Article 51 systemic risk classification"}
+ 3 more citations (full bibliography, deterministic workflow, actionable schema and crosswalks) included in the vault unlock — $0.01 via Skyfire / L402 / Direct Base USDC.
Access
- Discovery (free): /api/v1/nodes/eu-ai-act-article-51-gpai-systemic-risk-classification.json — 6-field metadata
- Vault (full node): /api/v1/vault/nodes/eu-ai-act-article-51-gpai-systemic-risk-classification.json — full 13-key payload, $0.01 USDC (L402/Skyfire/Direct Base)
- Canonical URL: https://bidda.com/intelligence/eu-ai-act-article-51-gpai-systemic-risk-classification
- Back to registry: Browse all 10,090 compliance nodes