Is European Democracy and Public Administration Ready for Artificial Intelligence? – Constitutional Discourse

In the 21st century, European Union law and constitutional law face new challenges. AI affects the exercise of public authority, democratic legitimacy and the institutional balance of the European Union. AI should not only be regulated, but public administration must be reformed to be better prepared for AI usage in the interest of European citizens. This post seeks to invite the reader to consider the ramifications of this issue and opens a debate regarding further constitutional and EU law analysis of AI tools in light of the upcoming AFCO report on institutional aspects of AI.
The Committee on Constitutional Affairs of the European Parliament is working on a report assessing the institutional aspects of Artificial Intelligence in European integration. The AFCO report is right in recognizing that AI governance is not only a technical, but a constitutional, EU law problem as well, affecting public authority, legitimacy and institutional balance.
AI challenges the foundations of governance. There is a growing tendency of applying AI systems within the public administrations. The AI approach of the EU needs to be in line with Article 2 TEU values. There is an outside dimension of the AI application in the report as well, as the EU’s global engagement needs to be more scrutinized. The text stresses that while there is strong geopolitical competition in the world, Parliament’s oversight in regulating and applying AI is essential.
The report not only states that „the EU’s governance architecture remains coherent, transparent, democratically legitimate and capable of responding effectively to emerging challenges”, it also stresses the respect for principles of subsidiarity, proportionality and the Member States’ constitutional identities, which implies that the EU and national level should cooperate in adequately addressing this issue.
The most important proposal within the text is the „possible evolution of the Treaties” in the medium to long term. This opens the debate about whether the founding Treaties should be amended to better address the effects of AI on European institutional setup and democracy.
In my opinion, most constitutional principles within the Member States’ legal orders and foundational principles are influenced by AI governance. The democratically elected MPs, MEPs, governments and the appointed EU senior officials all need to be aware of the foundational changes that AI brings to the public sphere and to our societies. There can be instances where AI could even affect the outcome of an election through disinformation on social media. The rule of law is affected as well, through possible interference into fundamental rights, like the protection of personal data, discrimination through distorted learning processes of AI, the problems of good administration, for example in case of automatic procedures. These problems were assessed by many before, but the question is, what should the focus of the AFCO report be.
Regarding structuring the work of the EP in this field, the external dimension and the EU public administration, good administration, proper regulation aspect should be divided into different reports. The effect of AI on CFSP, geopolitical considerations could be examined through a separate report of the foreign affairs committee. Another separate topic could be the division of regulatory tasks between the EU and national level. The focus of AFCO should be first and foremost to provide oversight over the introduction and application of AI tools into the everyday work of public administration, while making sure that the citizens of the European Union are protected from AI misuse. 
That being said, the right to good administration is especially vulnerable. There is a dire need for thorough legal assessment and parliamentary scrutiny of the application of AI by public administration in practice. Good administration can be easily breached by AI decisions, if the necessary hearing of one party’s position is lacking. Access to documents brings up further questions, as data used by the black box of AI cannot easily be accessible or even defining what a document is could pose challenges in this case. A black box system cannot adequately justify its decision as much of the AI processes are still not fully understood. These are just a couple of examples of a tendency that needs a great amount of attention from decision-makers.
The AFCO report sets out to analyze the European institutional aspects of AI, however, due to the complex range of issues it addresses, it only states general remarks and references general principles. The report could reflect on the relationship of AI with the principle of institutional balance, which is a core principle in the EU’s institutional set-up, representing the separation of power on the EU level. The EU institutions are obliged to act within their own scope of competence. AI gaining ground could lead to competition between EU institutions, one institution could be in a more favorable position having more information and data at hand. Also, the role of AI is different when executive power is applied, compared to legislative, or judicial functions. The role of the European Parliament and national parliaments should be to provide stronger parliamentary oversight over the executive branch, when using AI.
An important issue that is referenced indirectly is the relationship of national law and EU law and its AI-related implications. The principles of the respect for national identity, proportionality, subsidiarity all gain more importance as the national parliaments need to play a greater role in EU legislation regarding EU matters, as the national parliamentary oversight over AI usage of national authorities is essential as well.
As AI gains ground, the safeguarding of good administration requires more and more scrutiny from parliaments, while the executive power can become stronger and parliamentary accountability weaker. It goes back to the same conundrum as always with AI: will the European Union become dehumanized at the expense of European citizens, or will AI enhance the EU itself? Parliaments should adapt to these new challenges as soon as possible and a lot more work is needed from AFCO itself in this field. 
Árpád Lapu is an assistant research fellow at the Károli Gáspár University of the Reformed Church in Hungary. He was a policy adviser on constitutional issues at the European Parliament between 2019-2024. He worked as an adviser at the Ministry of Justice of Hungary (2017-2019) and the Ministry of European Union Affairs (2024-2026), conducting EU law, international law and comparative constitutional analyses. He has earned his JD at the Pázmány Péter Catholic University in Hungary, has a BA in international relations from the University of Szeged, and an MA in European and international administration from Andrássy Gyula German Speaking University in Budapest. He has completed an LLM in international law at the Catholic University of Louvain (UCLouvain).
Constitutional Discourse. All rights reserved. 2020-2026
Designed by GutiSoft.
We are using cookies to give you the best experience on our website.
You can find out more about which cookies we are using or switch them off in settings.
This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
Strictly Necessary Cookie should be enabled at all times so that we can save your preferences for cookie settings.

source