Artificial Intelligence

The CEN and CENELEC Joint Technical Committee 21 (JTC 21) is dedicated to developing European standards for Artificial Intelligence (AI), including harmonized standards in support of the EU AI Act. Established on June 1, 2021, JTC 21 unites over 300 experts from more than 20 countries, working collaboratively through five specialized working groups.

Objectives and Activities

JTC 21 aims to:

• Address European market and societal needs in AI standardization.
• Support the implementation of EU legislation, policies, and principles.
• Develop harmonized standards for legal conformity under the AI Act.
• Provide guidance to other technical committees on AI-related matters.
• Adopt international standards and develop homegrown European standards when needed.

Supporting the AI Act

The committee plays a pivotal role in implementing the EU AI Act by developing harmonized standards. A standardization request is issued by the commission to develop such standards (eNorm Platform). Once published in the Official Journal of the European Union (OJEU), these standards will provide companies with a legal presumption of conformity and ensure compliance with risk management, transparency, human oversight, cybersecurity, and quality assurance requirements.

Key Standards Under Development:

The full work programme of JTC 21 is available: here.

The key standards under development by JTC 21 supporting EU AI Act are as follow:

• AI Trustworthiness Framework.
• AI Risk Management: Addressing operational risks.
• AI Quality Management System: Ensuring robust processes for AI development.
• AI Conformity Assessment: Facilitating compliance verification.

These standards will be supplemented by a number of more specific standards which address specific sections of the standardization request. This includes standards on datasets, bias, computer vision, cybersecurity, robustness, logging and natural language processing.

Working Groups

The committee operates through five working groups:

1. Strategic Advisory Group (SAG): Focuses on overarching strategies and cross-sectoral challenges.
2. Operational Aspects: Addresses practical implementation issues.
3. Engineering Aspects: Concentrates on technical design and development.
4. Foundational and Societal Aspects: Examines ethical and societal impacts of AI.
5. Cybersecurity for AI Systems: Ensures secure AI deployment.

Global Collaboration and Future Goals

JTC 21 collaborates with the international standardization organizations ISO/IEC to adapt already existing global standards while addressing unique European needs.

For more details on ongoing projects or updates, visit JTC 21’s webpage and subscribe to the newsletter.

Given the complexity and wide range of topics within the scope of AI standardization, JTC 21 actively encourage participation from a wide array of stakeholders (e.g. SMEs, research institutions, industry representatives, regulators, and non-government organizations). This focus on consensus and transparency help ensure that the resulting standards are both technically sound and representative of all affected parties.

In particular, JTC 21/WG 1 has established a Task Group dedicated to Inclusiveness to raise awareness about JTC 21 activities and bring new stakeholders to the discussions.

The Task Group publishes a newsletter that is available to all interested parties, with the help of the European Trade Union Confederation (ETUC).

Feel free to join the LinkedIn Group on European Standards & Artificial Intelligence and register for the mailing list by emailing TG_Inclusiveness@etuc.org.

All previous editions can also be found here.