Views: 109
European experts’ community and officials, e.g. the former Italian PM and ECB President Mario Draghi’s views (basically expressed in a report in Autumn 2024) are of the opinion that recently adopted the EU-wide rules on AI legislation “should be paused” until all direct, indirect and potential “drawbacks are assessed”. However, present EU strategy is focused on broader AI models’ deployment as vital technologies in national economic growth, science and the EU’s competitiveness agenda.
Background
The EU progress in implementing the recommendations set out this fall at a special conference, that the Commission’s EU-wide digital transition process and legislation, generally, are subject to the inherent challenges threatening the whole structure of the European integration, in which the digital agenda occupies a vital place.
The Draghi Report already a year ago emphasised multiple risks to the European competitiveness’ issues due to looming global challenges and, in particular, to the Union’s falling behind position in strategic digital technologies, AI models, related software, hardware and cloud components. The slow pace of progress is, besides, mostly, due to insufficient level of private-public investments and excessive overregulation.
Generally, the AI’s strategic technologies’ agenda is based on two vital directions: a) rapidly expanding market in digital services, and b) emerging vital capacities in enhancing efficiency across numerous economic development and active corporate sectors. These dual aspects are underlining the EU-wide priority, which is presently focused on AI models as a category of technologies within the EU’s overarching competitiveness agenda”.
Eteris E. “EU-wide competitiveness: challenges and perspectives in Draghi report”, in:
https://www.integrin.dk/2024/09/09/eu-wide-competitiveness-challenges-and-perspectives-in-draghi-report/
Although the EU AI Act remains a vital legislation tool regulating AIs and digital transition, the adoption of the AI Continent Action Plan in April 2025 marked “a decisive shift” from primarily AI law towards a more growth-oriented approach: thus, i.e. the plan identified investment, infrastructure and skills as the essential pillars of the European AI leadership.
Furthermore, the AI action plan outlined some additional stimulating mechanisms, such as establishing of at least 15 operational “AI factories” in over half of the EU member states during 2025–2026, to consolidate availability of costly resources needed to facilitate deployment of sophisticated AI computing and data infrastructures.
Reference to: https://hcss.nl/news/the-draghi-report-revisited-artificial-intelligence-ai/
Positively, the Commission’s efforts over a couple of past year towards facilitating investment in AI technologies also stimulated creation of AI gigafactories aimed at establishing powerful digital training hubs. Thus, recent EU’s InvestAI program (February 2025), as a financial initiative, was aimed both to mobilize €200 billion in combined public and private investment in AI, including a new European fund of €20 billion for AI gigafactories. Creating such a large AI infrastructure is needed to support effective “collaborative development of the most complex AI models and to make Europe an AI continent”, noted Commission President in February 2025.
Source: European Commission. Press release “EU launches InvestAI initiative to mobilise €200 billion of investment in artificial intelligence”, in: https://luxembourg.representation.ec.europa.eu/actualites-et-evenements/actualites/eu-launches-investai-initiative-mobilise-eu200-billion-investment-artificial-intelligence-2025-02-11_en
“Postponing” the AI law
Mario Draghi suggested “radical” cuts in the European General Data Protection Regulation, GDPR and called for a pause on implementing some parts of the AI Act in order to foster regional AI development. The GDPR, as a comprehensive data privacy law, gives EU citizens additional control over their personal data; besides, the law requires organizations that handle such data to protect it. The regulation applies to companies inside and outside the EU if they offer goods or services based on the digital background connected to e.g. monitoring citizens’ behavior.
Note. The regulations, which is in effect since May 2018, replaced the outdated 1995 Data Protection Directive and aims to harmonize data privacy laws across all EU member states.
More on GDPR in: https://gdpr-info.eu/
In broader remarks, Mr. Draghi warned that Europe urgently needs “new speed, scale and intensity” in facing growing competition from China and the US, particularly in key tech sectors such as artificial intelligence.
More in: https://www.euractiv.com/news/draghi-calls-for-deep-cuts-to-privacy-rules-and-pause-on-high-risk-ai-act/
In his opinion of “postponing” the AI Act, Mr. Draghi, however, valuated existing and progressive “greater EU’s digital ambition”, saying that while the AI models represent key technologies, they “cannot be the sole focus for support”. Hence, he correctly stressed that the AI deployment depends on “at least four other technologies”: a) cloud, b) supercomputing facilities, c) cybersecurity, and d) connectivity as a vital component in developing efficient European AI systems.
As to the AI Act itself, Draghi seems calling it the legal instrument, which is going to be “put on posing” during the next stage of AI implementation in the EU, “until a better understanding the drawbacks”; he reminded on the Act’s provisions that refer to “high risks” for users of AI in areas like critical infrastructure.
At the same time, representatives from industrial sectors and the EU member states’ economic experts have been calling the EU to “stop the clock” on implementing the AI law for several months; the “citing delays” would be used to elaborate the needed compliance tools, additional digital-type structures and standards. [ ]
More in the “Draghi Report Revisited”, in: https://hcss.nl/draghi-report-revisited/
Suggestions
Draghi’s suggestions underlined the importance of collecting data in enabling optimal AI solutions; to his opinion, presently “the GenAI models require vast amounts of information to train”; hence, the European Commission must tackle obstacles to accessing data, while reducing the GDPR’s importance in “creating high levels of legal uncertainty for AI developers”.
Since 2023, when Italy’s privacy watchdog launched a probe into OpenAI’s ChatGPT, regulators across the EU have opened a number of investigations into “generative AI tools”, yet relatively few of these GDPR probes have been concluded, “as authorities wrestle with how to apply the bloc’s rules to cutting-edge technologies”, Draghi noted.
In line with these suggestions, the Commission intends “easing record-keeping” for smaller businesses (i.e. or companies with fewer than 750 employees), though “not ruling out a wider reform at a later stage”.
Source: https://www.euractiv.com/news/european-commission-looks-again-at-its-gdpr-privacy-law/