Defining the Future: The AI Act's Potential in equitably Safeguarding Fundamental Rights and Promoting AI Innovation

Abstract

The recent advancements in the realm of Artificial Intelligence (AI) feature a landscape filled with exciting prospects and lurking dangers in equal measure, necessitating a nuanced, human-centred regulatory approach. The EU is the first organisation to have undertaken this challenge with binding law through its AI Act. This paper delves into the complex task of this founding stone of EU AI law to balance promoting innovation and safeguarding fundamental rights. It addresses both the Act’s pre-history – encompassing the relevant policy documents before the European Commission's proposal – and its history, which includes the negotiations and amendments leading up to the final trilogue session on the 6th of December 2023 and the consequential provisional agreement, followed by its approval by COREPER and the IMCO and LIBE committees and its final upvoting in the European Parliament on the 13th of March 2024. Keeping track of the Act’s legislative journey, including the context in which said journey evolved, shall be crucial in order to comprehensively understand, interpret, and apply what is expected to be one of Europe’s most impactful pieces of legislation to date. The paper aims to bridge a gap in scholarly discussion around the AI Act, which has so far focused solely on evaluating fundamental rights protection or assessing cost-efficiency, providing a practical approach that consolidates the legislative process and appraises the AI Act’s overall potential in protecting fundamental rights, while equitably maintaining AI innovation. The research question it seeks to answer is: How can the AI Act promise sufficient fundamental rights protection without compromising AI innovation and overburdening enterprises? The paper is structured to provide a comprehensive analysis of the Act in view of answering the research question. The Overview section (II.) will give the historical background of AI and track the main policy documents that led to the AI Act’s proposal offering a source also for retrospective research in future. It will also identify specific key points of the Commission’s original proposal that call for examination and establish its connection to fundamental rights protection, in view of being presented as a product safety regulation. The section on the Legislative State of the Art and Remaining Pitfalls (III.) will present the fruits the legislative process has yielded up until the confirmation of the agreed final draft by the IMCO and LIBE committees on the 13th of February 2024 and its subsequent approval in the Parliament a month later. It will enumerate the main points introduced with the negotiated amendments to then identify still existing potential pitfalls with regards to the Act’s scope, enforcement mechanism and future proofness, and how this creates weaknesses for innovation and fundamental rights. The Balancing section (IV.) offers suggestions to counter the weaknesses that were identified in the previous part, uncovering the balancing point between fundamental rights protection and innovation. Finally, the paper will conclude (V.) by summarising the findings and the answer to the research question while leaving room for future discussion.