EU · binding regulation
EU AI Act
EU-AIA-2024
Treatment of Foundation Models / GPAI
governsArts. 51-55 (general-purpose AI + systemic risk)
Primary source: Regulation (EU) 2024/1689
Comparative jurisdictional memo · Foundation Models / GPAI
EU AI Act vs Executive Order 14110 on Safe, Secure, Trustworthy AI
Verdict: convergesDownload PDF
US · executive order
Executive Order 14110 on Safe, Secure, Trustworthy AI
US-EO-14110
Treatment of Foundation Models / GPAI
governs§4.2(a) — Defense Production Act reporting
Primary source: Exec. Order No. 14110, 88 Fed. Reg. 75191 (Nov. 1, 2023)
What this comparison shows
On the topic of Foundation Models / GPAI, EU AI Act is classified as governs and Executive Order 14110 on Safe, Secure, Trustworthy AI as governs. The editorial verdict is converges. The two regimes take the same position; this is a candidate for cross-jurisdictional alignment in industry guidance, voluntary codes, and harmonisation tracks.
Contested question: Does the foundation-model category map to a coherent capability tier, or is it a regulatory convenience? Compute-threshold vs behavioural-threshold debate is unresolved across EU/US/China.
Bibliography
- Regulation (EU) 2024/1689 — EU AI Act.
- Exec. Order No. 14110, 88 Fed. Reg. 75191 (Nov. 1, 2023) — Executive Order 14110 on Safe, Secure, Trustworthy AI.
- Policy Window — Foundation Models / GPAI (cross-jurisdiction topic article with full 26-instrument coverage matrix).