European Union: Issued European Commission request to develop harmonised standards for the implementation of the AI Act

On 24 October 2024, the European Commission issued a formal standardisation request to the European standardisation organisations, CEN and CENELEC, to draft harmonised standards for the implementation of the Artificial Intelligence (AI) Act. Adopted in August 2024, the AI Act sets out requirements for high-risk AI systems, ensuring they respect safety, health, and fundamental rights. Following a transition period of two to three years, beginning in August 2026, high-risk AI systems will be required to comply with specific criteria, including risk management, data quality and governance, logging and traceability, technical documentation, transparency, human oversight, accuracy, robustness, and cybersecurity. The request aims to establish a legal presumption of conformity for AI systems developed according to these forthcoming standards, thus facilitating their market access and promoting safe and trustworthy AI across the European Union. By defining the essential requirements for high-risk AI systems, the Commission seeks to create a framework that supports compliance with the AI Act while involving various stakeholders, including small and medium-sized enterprises.