On the 1st of August 2024, European Union’s Regulation 2024/1689 (Artificial Intelligence Act) entered into force, laying down harmonized rules regarding Artificial Intelligence matters.
Amongst others, the Act defines in Chapter VII: Governance (articles 64-70) the establishment of its five governing and supervisory bodies;
1. Artificial Intelligence Office (Article 64):
The “European AI Office” is established with the support of EU Member States as the central hub for AI expertise in the Union. It is tasked with the implementation of the AI Act, ensuring the development of “trustworthy AI” while mitigating risks and promoting legal certainty.
2. Artificial Intelligence Board (Articles 65-66):
The “European Artificial Intelligence Board (AI Board)” is established, comprising high-level representatives from the European Commission and all EU Member States. The Board’s role is to assist and advise the Commission and Member States in applying the regulation effectively by sharing best practices, issuing recommendations, and collaborating with other EU institutions and national authorities.
3. Advisory Forum (Article 67):
An Advisory Forum is established, entrusted with the duty to provide the Board and the Commission with technical expertise and advice.
4. Scientific Panel of independent Experts (Articles 68-69):
A scientific panel of independent experts is established to support the enforcement activities under the AI Regulation.
5. National Competent Authorities and Single Points of Contact (Article 70):
Each Member State shall establish or designate as national competent authorities at least one “notifying authority” and at least one “market surveillance authority” for the purposes of the Regulation.
Even though the AI Act becomes effective on August 2 2026, the provisions related to establishing the governing bodies come into effect one year earlier, on August 2 2025 (Article 113), allowing the EU Member States an ample period of time to be prepared.
The governance framework established by the AI Act bears notable similarities to the supervisory mechanism of the General Data Protection Regulation (EU 2016/679 – GDPR), reflecting the EU’s commitment to a unified regulatory approach. Specifically, both regulations emphasize interdisciplinary coordination by creating multiple governing bodies and implementing a structured oversight system.
Following the successful monitoring and supervisory model of the GDPR, which has effectively guided organizations towards better practices, the AI Act is designed to achieve comparable outcomes in the area of AI governance.
However, even though the AI Act’s oversight framework draws on the GDPR’s governance model, it may raise concerns about excessive complexity. Moreover, while the intention behind is to ensure robust and comprehensive oversight, this multi – leveled governance system could lead to managerial inefficiencies, overlapping responsibilities and jurisdictional conflicts.
Therefore, establishing a balance between thorough oversight and operational efficiency is essential in order to achieve optimal results.
Ultimately, it remains to be seen whether the governing bodies established under the AI Act will function as intended or whether they might bring unforeseen challenges alongside the solutions they aim to provide.
Edited by Michales Kamposos