The European Parliament's ECR group has tabled a single amendment (131/rev) to the AI Act simplification package that replaces the conditional application of core obligations for high-risk AI systems with a fixed deadline of 2 August 2028. The amendment, submitted by MEPs Piotr Müller, Stefano Cavedagna, and Kosma Złotowski, targets the transitional provisions of the legislative report (A10-0073/2026) on simplifying the implementation of harmonised AI rules.
The amendment removes the original text's mechanism that tied the application of Chapter III (Sections 1-3) obligations—covering risk management, data governance, transparency, and human oversight—to a future Commission decision confirming that compliance support measures are available. Instead, it sets a definitive date of 2 August 2028, exactly six months after the regulation's proposed entry into force on 2 February 2028. The amendment explicitly excludes Article 6(5) on classification rules for high-risk AI systems, which would still apply from the earlier general date of 2 August 2027.
Policy orientations and trade-offs
The ECR's proposal prioritises legal certainty and predictability for providers and deployers of high-risk AI systems over the original phased approach that aimed to ensure practical readiness. By removing the Commission's conditional trigger, the amendment forces all stakeholders to prepare for a known deadline regardless of the Commission's administrative progress on standards, guidelines, or testing facilities. This reduces regulatory limbo but risks imposing obligations before the necessary compliance infrastructure is fully in place. The trade-off is between a fixed, business-friendly timeline and a more cautious, readiness-based approach.
Impact on stakeholders
- Providers of high-risk AI systems: gain a clear compliance deadline, enabling investment planning and resource allocation, but face pressure to meet obligations without guaranteed support measures.
- National supervisory authorities: must be ready to enforce by August 2028, potentially without full Commission guidance or harmonised standards.
- EU Commission: loses discretion to delay obligations if support measures are not ready, shifting responsibility to member states and industry.
- Deployers of high-risk AI systems: benefit from legal certainty but may face earlier compliance costs if support tools are incomplete.
Institutional follow-up
The amendment will be considered as part of the plenary vote on the AI Act simplification package. The Council's position and subsequent trilogue negotiations will determine whether the fixed deadline or the conditional approach prevails. The European Parliament's lead committee (IMCO/LIBE) had previously supported the original conditional text, setting up a potential floor debate between legal certainty and phased readiness.