TLDR: The European Union is reportedly considering significant adjustments and new grace periods for its Artificial Intelligence Act, aiming to ease compliance burdens for companies. A draft proposal, seen by The Wall Street Journal, suggests offering extensions for generative AI systems to meet transparency obligations and delaying fines for certain requirements until August 2027. These changes are part of a broader effort to simplify digital regulations and address concerns from the tech industry and member states.
Brussels, November 7, 2025 – The European Union is poised to introduce modifications and extended compliance timelines for its groundbreaking Artificial Intelligence Act, according to a draft proposal obtained by The Wall Street Journal. These prospective changes are designed to streamline adherence for businesses and alleviate concerns raised by the tech industry and various member states regarding the law’s implementation. The European Commission is expected to formally unveil this package of measures on November 19, though the specifics could still evolve.
The proposed tweaks are part of a wider initiative by Brussels to simplify the bloc’s digital regulatory framework. A key element of the draft includes offering a grace period to companies whose generative AI systems are already operational, allowing them more time to comply with obligations such as watermarking to enhance transparency. Furthermore, the proposal suggests delaying the imposition of fines for non-compliance with certain transparency requirements by one year, pushing the deadline to August 2027. This extension is intended to “provide sufficient time for adaptation of providers and deployers of AI systems” to meet the new rules.
European Commission digital spokesman Thomas Regnier acknowledged the ongoing dialogue, stating, “We are hearing concerns from the industry and from our member states.” He indicated that the upcoming package of measures would serve as “the appropriate framework to address some of these concerns.” Regnier also reaffirmed the EU’s unwavering commitment to its regulatory sovereignty, noting that while EU officials have been in continuous contact with U.S. counterparts, “It’s not for a third country to decide how we legislate.” He emphasized, “We are fully behind the AI Act, and no pressure from anywhere will impact us.”
Also Read:
- EU Appoints Experts to Lead Generative AI Content Labeling Code of Practice
- Navigating the Legal Landscape: AI, Copyright, and Litigation in the EU and UK
The AI Act, which officially entered into force last year, has a staggered implementation schedule, with various obligations kicking in over several years. The EU has faced considerable pressure from tech companies and lobby groups, who argue that the regulations could create excessive bureaucracy and hinder innovation. The law has also been a point of contention with Washington, as U.S. officials have expressed concerns that the rules could burden American tech enterprises. The Commission has previously stated its commitment to an “innovation-friendly implementation of the AI Act,” and these proposed adjustments appear to reflect that stance.
