The European approach to regulating AI through technical standards

Mélanie Gornet (1, 2, 3, 4)

1 IP Paris - Institut Polytechnique de Paris
2 SES - Département Sciences Economiques et Sociales
3 NOS - Numérique, Organisation et Société
4 Télécom Paris

European institutions are currently finalizing the AI Act, a new regulation on artificial intelligence. The AI Act will require manufacturers of high-risk AI systems to affix a European Conformity (CE) mark on their products. The CE mark will show compliance with the essential requirements of the AI Act and permit AI products to circulate freely in Europe. For the mark to be issued, manufacturers will have to apply harmonized technical standards (HSs). The CE mark and HSs are long-established European regulatory tools to deal with product safety, and already apply to a wide range of products, like toys, machinery or medical devices. To date, however, they have never been used to attest to compliance with fundamental rights, yet this is part of the goal of the AI Act. This article examines the role of HSs and CE marking in the AI Act, and in particular how these product safety regulatory techniques have been expanded to cover protection of fundamental rights. This article further examines the June 22, 2023 opinion of the CJEU Advocate General in the Public.Resource.Org case appeal, which characterizes HSs as regulatory instruments under EU law, and how this qualification raises questions on democratic processes and transparency in standardization organizations. Finally, this article examines issues raised by the EU's choice, in the AI Act, to cover fundamental rights through HSs and the CE mark. This article concludes that unlike compliance with product safety, compliance with fundamental rights cannot be attested to by technical standards and CE marking because violations of rights are too context-specific and require a judicial determination. However, technical standards have an important role to play in encouraging best practices in AI governance, which take protection of fundamental rights into account.

Domaines: Droit; Intelligence artificielle [cs.AI]

Mélanie Gornet. The European approach to regulating AI through technical standards. 2023. ⟨hal-04254949⟩