EU Strikes Deal to Delay AI Act Compliance Deadlines, Easing Pressure on Businesses

EU Agrees to Softer AI Act Timeline

European Union lawmakers reached a provisional agreement early Thursday to push back the most stringent deadlines under the bloc’s AI Act, giving companies more time to prepare for high-risk compliance. The deal, struck between negotiators for the European Parliament and the European Council, shifts the original August 2, 2026 deadline for high-risk AI systems to December 2, 2027 for stand-alone systems and August 2, 2028 for AI used in products covered by sectoral safety rules.

“Today’s agreement on the AI Act significantly supports our companies by reducing recurring administrative costs,” said Marilena Raouna, Cyprus’s deputy minister for European affairs, in a statement from the Council. The deal still requires formal adoption by both the Parliament and the Council before it becomes law, with legislators aiming to complete the process before the original deadline takes effect.

Background

The AI Act, first proposed by the European Commission in 2021, aims to regulate artificial intelligence based on risk levels. High-risk applications, such as those used in critical infrastructure or law enforcement, faced the strictest rules and the earliest compliance date—August 2, 2026.

Previous negotiations collapsed nine days before this deal, highlighting tensions between member states wanting to foster innovation and those prioritizing safety. The provisional agreement addresses industry concerns about overlapping requirements and unreasonable timelines.

What This Means

Businesses now have up to two additional years to align with high-risk AI obligations, reducing immediate pressure on sectors like healthcare, automotive, and manufacturing. The deal also narrows the definition of “safety component” under the AI Act, meaning AI features that only assist users or improve performance will not automatically be treated as high-risk—provided their failure doesn’t create health or safety risks.

“With this agreement, we show that politics can move just as quickly as technology,” said Arba Kokalari, the Parliament’s co-rapporteur for the Internal Market and Consumer Protection committee. However, the original deadline remains in effect until formal adoption, so companies should continue preparations.

Key Changes Under the Deal

• High-risk AI deadlines: Stand-alone systems by Dec 2, 2027; AI in regulated products by Aug 2, 2028.
• Overlapping rules removed: AI in machinery products will now follow only sectoral safety rules, with health and safety safeguards.
• Regulatory sandboxes: Deadline for member states to set up AI regulatory sandboxes pushed back by a year to Aug 2, 2027.
• Watermarking obligations: Requirements for AI-generated content will apply earlier than proposed, from Dec 2, 2026 instead of Feb 2, 2027.
• Mid-size firm exemptions: Small mid-cap companies now qualify for exemptions previously limited to SMEs.
• AI Office supervision: Central oversight for general-purpose AI systems, while national authorities retain responsibilities in law enforcement, border management, and financial institutions.

The co-legislators also agreed on a mechanism to resolve overlaps between the AI Act and existing sectoral laws for medical devices, toys, lifts, machinery, and watercraft. This ensures clarity for industries already subject to strict regulations.

“This provisional agreement strikes a balance between innovation and safety,” the Council statement added. The next step is formal adoption by the European Parliament and Council. Until then, the original August 2, 2026 deadline remains legally binding.

For more details, see the Background section or review the full legislative text once published.