Fri. Nov 22nd, 2024
ai, artificial intelligence, sci-fi
How to make AI safe and trustworthy? Photo by tungnguyen0905 on Pixabay

Today, the Commission’s Artificial Intelligence (AI) Office is launching a targeted stakeholders consultation process on the future guidelines on the AI system definition and the implementation of AI practices that pose unacceptable risks under the AI Act.

The guidelines will help national competent authorities as well as providers and deployers in complying with the AI Act’s ruleson such AI practices ahead of the application of the relevant provisions on 2 February 2025.

The AI Office invites stakeholders, includingAI systems providers, businesses, national authorities, academia, research institutions and civil societyto submit their input. The contributions received will feed into the Commission’s guidelines on the definition of AI system and prohibited AI practices under the AI Act, to be published in early 2025.

The legal concepts regarding the AI system definition and prohibited AI practices are established in the AI Act. This consultation seeks additional practical examples from stakeholders to feed into the guidelines and provide further clarity on practical aspects and use cases.

Participants can find more information on the consultation below, which will be active later today. The consultation will remain open for 4 weeks, until 11 December 2024.

More information:

 


Multi-stakeholder consultation for Commission guidelines on the application of the definition of an AI system and the prohibited AI practices established in the AI Act

Disclaimer: This document is a working document for consultation and does not prejudge the final decision that the Commission may take on the final guidelines. The responses to this consultation paper will provide important input to the Commission when preparing the guidelines.

The European AI Office is launching this multi-stakeholder consultation on the application of the definition of an AI system and the prohibited AI practices established in the AI Act. This consultation is targeted to stakeholders of different categories, including providers and deployers of AI systems such as businesses, authorities (including local public authorities) and other organisations, academia and research institutions, trade unions and other workers’ representatives, civil society organisations, public supervisory authorities, and the general public.

As not all questions may be relevant for all stakeholders, respondents may reply only to the section(s) and the questions they consider relevant. Respondents are encouraged to provide explanations and concrete cases as part of their responses to support the practical usefulness of the guidelines.

The targeted consultation is available in English only and will be open for 4 weeks starting on 13 November until 11 December 2024. We strongly encourage early submissions.

The questionnaire for this consultation is structured along 2 sections with several questions.

1. Questions in relation to the definition of an AI system

2. Questions in relation to prohibited AI practices

We welcome collective answers from organisations. You have the option to indicate if you a submitting such a collective answer in the end of the first section and identify the organisations on whose behalf the submission is made.

We welcome full or partial replies from all respondents based on their expertise and perspective.

All contributions to this consultation may be made publicly available. Therefore, please do not share any confidential information in your contribution. Individuals can request to have personal information removed from their contribution.

The Commission may publish a summary of the results of the consultation. In that case, results will be based on aggregated data and respondents will not be directly quoted.

Please allow enough time to submit your application before the deadline to avoid any issues. In case you experience technical problems which prevent you from submitting your application within the deadline, please take screenshots of the issue and the time it occurred.

In case you face any technical difficulties or would like to ask a question, please contact: CNECT-AIOFFICE@ec.europa.eu

General Introduction

The Artificial Intelligence Act (Regulation (EU) 2024/1689, hereinafter ‘the AI Act’), which entered into force on 1 August 2024, improves the internal market by laying down harmonised rules for trustworthy and human-centric Artificial Intelligence (AI) in the EU (Article 1 AI Act). It aims to promote innovation and uptake of AI, while ensuring a high level of protection of health, safety and fundamental rights, including democracy and the rule of law.

The AI Act establishes a common definition of an AI system, aligned with the OECD definition (OECD Recommendation on Artificial Intelligence (OECD/LEGAL/0449, 2019, amended 2023)), as a central element of the scope of the AI Act (Article 3(1) AI Act and Recital 12). The AI Act follows a risk-based approach to regulating AI systems, by classifying such systems into different risk categories. One of which are the prohibited AI practices covering AI systems posing unacceptable risks to fundamental rights and European values (Article 5 AI Act).

Pursuant to Article 96(1) AI Act, the Commission must develop guidelines on the practical implementation of the Regulation, inter alia, on the prohibited AI practices referred to in Article 5 AI Act and the application of the definition of an AI system as set out in Article 3(1).

The purpose of the present targeted stakeholder consultation is to collect input from a wide range of stakeholders on concrete examples of AI systems and issues with the practical application of the relevant AI Act provisions that could be clarified in the Commission’s guidelines on the definition of an ‘AI system’ as well as guidelines on the prohibited AI practices. The definitions and prohibitions are applicable six months after the entry into force of the AI Act, as from 2 February 2025. The input from this consultation will feed into the Commission guidelines to be adopted in early 2025. It should be noted that the legal concepts in relation to the AI system definition and the prohibitions are already set out in the AI Act. The Commission launches the present consultation to seek additional practical examples from stakeholders to feed into the guidelines and provide further clarity on practical aspects and use cases.

The objective of the guidelines is to provide consistent interpretation and practical guidance to assist competent authorities in their enforcement actions as well as providers and deployers subject to the AI Act in their compliance actions with a view to ensuring consistent, effective and uniform application of the prohibitions and understanding of what constitutes an AI system within the scope of the AI Act.

Take the questions: Consultation questions

 

Forward to your friends