The Role of ETSI in the EU’s Regulation and Governance of Artificial Intelligence

Marta Cantero Gamito
DOI: https://doi.org/10.2139/ssrn.4770324
2024-01-01
SSRN Electronic Journal
Abstract:As artificial intelligence (AI) technologies rapidly advance, they bring about important societal implications involving privacy, fairness, non-discrimination, and other relevant ethical considerations. Legislators and policymakers are joined by a common drive to provide legislative solutions and regulatory frameworks that guarantee that the ongoing integration of AI systems into society is consistent with fundamental rights and democratic values. This article explores the significant role that standardisation plays in this regulatory process and how it impacts the regulation and governance of AI within the European Union (EU). In particular, the paper provides a critical analysis of the regulatory approach adopted by the EU legislator for the AI Act, which delegates the definition of essential requirements for high-risk AI systems to harmonised standards, underlining the significance of standardisation in ensuring technical feasibility and compliance with EU laws and values. At the forefront of this discussion, there is the increasing influence of AI-related standardisation across social, economic, and geopolitical domains, with a particular focus on the crucial role played by Standard Developing Organisations (SDOs) in the regulatory and governance processes. This paper contributes to the legal scholarship by critically analysing the regulatory approach chosen for the EU’s AI Act, contesting the adequacy of the New Legislative Framework for AI governance, and arguing that the reliance on harmonised standards risks undermining democratic accountability and fails to sufficiently safeguard fundamental rights without a more inclusive and transparent standard-setting process. The article focuses on the exclusion of the European Telecommunications Standards Institute (ETSI) from the European Commission’s standardisation request in support of the AI Act and assesses its potential impact on EU law-making and regulatory consistency. Ultimately, the analysis aims to contribute to understanding standardisation dynamics, offering insights into its profound implications for AI governance and the broader digital sphere.
What problem does this paper attempt to address?