Work Programme - AI Standards in Manufacturing
European Standardisation body CEN/CLC/JTC 21
The European Standardization organizations CEN and CENELEC are tasked with developing a broad range of standards commissioned by the EU. Below is an overview of their work program for the upcoming two years.
Although these standards are still under development, they are expected to play a key role in ensuring presumed conformity with the forthcoming AI Act.
This information was last updated in November 2024. For more detailed information on their Artificial Intelligence work program, check out this AIRISE page, or their official webpage.
| Name | Description | Status |
|---|---|---|
| prCEN/CLC/TR 17894 | This document reviews current tools and methods for assessing whether AI systems align with key industry standards. It covers both general and industry-specific perspectives, focusing on evaluating products, services, processes, and management systems. Key areas include identifying compliance gaps and addressing unique AI challenges like data quality and software engineering reliability. Aligned with European policies, this guidance helps SMEs ensure their AI applications meet essential requirements, supporting trustworthy and compliant AI adoption. | Approval |
|
prCEN/CLC/TR XXX (WI=JT021009) Checklist for AI Risk Management | This document offers a checklist of risk criteria for evaluating AI systems, covering potential risks, harms, and countermeasures. While it doesn’t provide a specific method, it outlines criteria and possible contingency plans, with detailed examples included in the annex. Applicable to all types of organizations, including SMEs, this guidance helps smaller enterprises identify and address AI-related risks in a structured, practical way. | Preliminary |
|
FprCEN/CLC/TR 18145 Environmentally sustainable Artificial Intelligence | The document outlines a framework to help you quantify the environmental impact of AI and improve its sustainability and efficiency. It summarizes how AI can optimize energy systems, reduce direct (Scope 1) and indirect (Scope 2) emissions, and assess lifecycle emissions (Scope 3). It also highlights AI’s potential to measure avoided emissions (Scope 4) and maximize carbon savings in your operations. | Approval |
|
prCEN/CLC/TR XXXX (WI=JT021002) | Artificial Intelligence - Overview of Al tasks and functionalities related to natural language processing: This document is still under drafting in collaboration with ISO/IEC JTC 1/SC 42. | Under drafting, no info |
|
prEN ISO/IEC 12792 Transparency taxonomy of AI systems | This document defines a taxonomy of information elements to assist AI stakeholders with identifying and addressing the needs for transparency of AI systems. The document describes the semantics of the information elements and their relevance to the various objectives of different AI stakeholders. This document uses a horizontal approach and is applicable to any kind of organization and application involving AI. | Approval |
|
prEN ISO/IEC 23282 Evaluation methods for accurate natural language processing systems | This document helps you evaluate natural language processing systems by measuring the quality of their results to assess functional suitability. It defines evaluation methods, provides guidance on their selection and implementation, and explains how to interpret results. It covers both quantitative metrics and other evaluation approaches, with requirements for metrics implementation and technical resources used in the process. | Drafting |
|
prEN ISO/IEC 24029-2 Assessing Robustness of Neural Networks in AI | This document provides methodology for the use of formal methods to assess robustness properties of neural networks. The document focuses on how to select, apply, and manage formal methods to prove robustness properties. | Rev, already uploaded |
|
prEN ISO/IEC 42001 AI Management System | No information available | Preliminary |
|
prEN XXX / (WI=JT021019) Competence Requirements for AI Ethicists | This document provides a framework outlining the knowledge, skills, and attitudes required for AI ethicists, focusing on European values and fundamental rights. It categorizes competencies needed for the role and offers recommendations to ensure effective performance. This standard supports organizations—including SMEs—in integrating the role of AI ethicists, promoting responsible AI practices across diverse sectors. | -- |
|
prEN ISO/IEC 24970 AI system logging | This document introduces requirements and technical formats for logging AI systems in line with the AI Act’s record-keeping rules. It defines machine-readable log formats, including syntactical and semantic schemas, and provides recommendations for transmitting and storing data for post-market monitoring. Existing standards are referenced where possible to ensure consistency. SMEs can use this to streamline compliance and maintain transparent, structured records for AI systems. | Draft |
|
prEN XXX Data terms measures and bias requirements | This is still a preliminary work item. However, the abstract/scope this document present that the purpose of this proposed European Norm is to define the data terms measurements and bias requirements, addressing shortcomings and monitoring of the data processed by AI systems in the context of avoiding unwanted bias and proxy discrimination. | Cant find it on website |
|
prEN XXXXX AI-enhanced nudging | This document defines standards and guidelines for AI-enhanced nudging mechanisms—digital nudges shaped by AI systems. It helps you align with regulations like GDPR by providing requirements for designing responsible nudges, along with use cases, best practices, and auditing indicators. These tools guide you in assessing and improving your AI-powered nudging mechanisms. By applying this framework, you can ensure your nudges are ethical, compliant, and effective in engaging users. | Under Drafting |
| prCEN/TS (WI=JT021033) | This document provides guidance for upskilling on AI ethics and social concerns within your organization. It complements existing standards by focusing on building basic capabilities and fostering an ethical culture at both individual and organizational levels. It emphasizes continuous learning programs tailored for employees, teams, and leadership, excluding AI ethics professionals. By following this, you can ensure ethical AI practices are integrated across all roles, regardless of your organization's AI involvement. | Preliminary |
| Guidance for upskilling organisations on AI ethics and social concerns | No Information available | -- |
|
prCEN/TS (WI=JT021035) Sustainable Artificial Intelligence | This document outlines principles and a framework for measuring and reducing the environmental impact of AI systems throughout their lifecycle. It provides a harmonized calculation method, reporting guidelines, and best practices for impact reduction. It applies to AI developers, users, and all value chain actors involved with AI systems and services. By adopting this framework, you can minimize your AI’s environmental footprint while aligning with sustainability goals. | Prelimninary |
|
prCEN/TS (ZI=JT021034) Ethical issues handling in AI System LC |
This document provides tools and guidelines for addressing social and ethical concerns throughout the AI system lifecycle, from identifying issues to decision-making. It includes a list of tools with minimum requirements to help you implement them effectively, independent of specific ethical frameworks. While applicable to all organizations, it primarily supports AI producers and providers in managing societal and ethical considerations. By using this, you can ensure responsible AI practices are embedded in your processes and governance. | Preliminary |
|
prEN ISO/IEC 5259-1 (WI=JT021040) Data Quality : Overview, terminology, examples | This document provides the means for understanding and associating the individual documents of the ISO/IEC “Artificial intelligence — Data quality for analytics and ML” series and is the foundation for conceptual understanding of data quality for analytics and machine learning. It also discusses associated technologies and examples (e.g. use cases and usage scenarios | -- |
|
prEN ISO/IEC 5259-1 (WI=JT021040) Data Quality : Overview, terminology, examples | No information available | -- |