EU Reportedly Watering Down AI Act, but Denies U.S. Influenceh2>
The European Commission is considering moderating its landmark b>AI lawsb> amid reported pressure from U.S. tech companies and the Trump administration. This potential adjustment aims to simplify the EU’s digital rule book, including the b>AI Actb>, to reduce administrative burdens on European companies.p>
Statements from the European Commissionh3>
During a press briefing, Commission spokesperson b>Thomas Regnierb> emphasized, “Let me be extremely clear about one point: the U.S. will not decide on our simplification agenda. It’s our sovereign right to legislate.” He reassured that the Commission is not backing away from its digital legislation, maintaining complete independence in its legislative processes.p>
Additionally, b>Paula Pinhob>, chief spokesperson of the European Commission, noted that stakeholders show interest and share their views during the legislative preparation. However, she clarified that the extent to which those views are incorporated remains a different question.p>
Background Discussionsh3>
According to an interview with an unnamed senior EU official, the EU has been in discussions with the Trump administration regarding adjustments to the AI Act and other digital regulations as part of a streamlining effort. This follows the European Commission’s ongoing initiatives to simplify its b>datab>, b>cybersecurityb>, and b>AI legislationb> after a public call for evidence in September.p>
Goals of the Simplification Agendah3>
The simplification agenda aims to boost competitiveness while reducing administrative burdens on EU companies, all while upholding standards and maintaining values. Major tech companies like b>Metab> and b>Alphabetb> have previously contested the EU’s AI regulations, arguing that these rules restrict access to new and innovative services.p>
Potential Changes to the AI Acth3>
The b>AI Actb> is designed to ban risk systems deemed “unacceptable,” such as b>social scoringb>, and enforce strict requirements on high-risk areas including b>healthcareb>, b>employmentb>, and b>policingb>. These regulations apply to companies within the EU and any organization providing AI products or services to Europeans. It mandates rigorous b>transparency standardsb> for international companies, imposing substantial fines for breaches.p>
Draft legislation indicates that providers of b>generative AI systemsb> already in the market before the implementation date may secure a one-year grace period to adapt their practices. Furthermore, the Commission plans to delay fines for violations of the updated AI transparency rules until b>August 2027b>, allowing sufficient time for adaptation.p>
The draft also considers removing the compliance burden from companies by centralizing enforcement through its own AI office. Previously, the Commission rejected calls from major tech firms, including b>ASMLb>, b>Mercedes-Benzb>, and b>United Internetb>, to pause the EU AI Act, emphasizing that the complexity introduced by these regulations could hinder AI development on the continent.p>
