On Tuesday, November 14th, at the I-Com premises in Brussels, the network presented its joint paper on the EU Artificial Intelligence Act (AI Act) as the primary outcome of its activities in the second semester of 2023.
The paper was unveiled at the conference titled “AI and the Future of EU Competitiveness: The Road Ahead.” This event provided a valuable opportunity to discuss the current state of the EU AI Act.
With the aim of contributing to the debate from a Southern European perspective, the paper initiated discussions on the latest developments of the dossier preceding the agreement that negotiators aspire to reach by the end of 2023. It also addressed divisive topics, such as the use of copyrighted content by AI and biometric surveillance.
Chapter 1, in addition to providing an overview of four major international initiatives and approaches to AI regulation, centers on the EU AI Act. It emphasizes that critical issues still require resolution, including those discussed in the trilogue and the terms and role of human oversight. The chapter underscores that the EU AI Regulation will need regular updates due to the ever-changing nature of AI and the potential unintended legal fragmentation resulting from the AI Act’s subsidiarity provisions (penalties, high-risk classification). It also outlines various areas where trilogue negotiations must establish consensus among the Commission, Parliament, and Council.
Chapter 2 analyzes the economic impact of generative AI, with a specific focus on Southern European countries such as Greece, Italy, Portugal, and Spain. It provides a global overview of the generative AI landscape and discusses potential challenges and strategies for addressing them.
Chapter 3 explores the significant differences between generative AI and traditional AI, as well as the leading countries in the AI race in terms of private investments, patents, and publications. The analysis highlights the ongoing competition between the United States and China, with both countries consistently leading the rankings. However, it also notes that many widely-used generative AI projects have a global reach as they are open-source.
Chapter 4 examines the current low adoption of AI technologies, emphasizing the opportunity for a mindful deployment of these technologies to maximize their benefits while mitigating potential pitfalls. Through a comprehensive SWOT analysis, the authors highlight the steady increase in relevant research in the network’s countries over the last two decades and a significant surge in investments in AI startups in the last three years.