The following is an adapted version of an article written by Anna Taraczközi, a research fellow at the Europe Strategy Institute of the University of Public Service, originally published in Hungarian on the Five Minutes Europe blog of Ludovika.hu.
The RAISE initiative, launched at the Dublin summit, aims to connect the continent’s AI research resources, but sovereignty and trust are also at stake.
Europe has long prided itself on scientific excellence, but it is increasingly concerned that, in the age of artificial intelligence (AI), it will fall behind in its application. The Commission responded to this tension when it launched RAISE—the Resource for AI Science in Europe initiative—at the AI in Science summit in Dublin on 15 October. The goal is ambitious: to bring the continent’s scattered AI research resources—from computing capacity to data to talent—under a single common framework so that European scientists do not fall behind their American and Chinese competitors.
RAISE is not a new building or office, but a virtual, pan-European research institute that brings together member states and the private sector’s existing capacities. A central secretariat will coordinate the thematic networks of excellence so that results from one scientific field can benefit others. At the same time, three platforms—the RAISE Hub, the Forum, and the Observatory—will organize access, collaboration, and monitoring. The logic is familiar: for years, the idea of ‘AI-CERN’—a joint European institution that would give research autonomy and scale—has been recurring. RAISE is a distributed, networked version of this, not built in a single location but connecting existing strengths.
Funding is also flowing behind it, albeit cautiously for now. The initiative is launching with a pilot programme of about 107 million euros from Horizon Europe; 58 million will fund networks of excellence and doctoral networks, and an additional 600 million euros will provide researchers with dedicated access to the AI gigafactories currently under construction. According to the broader plan—the Commission’s AI in Science strategy—annual AI spending under Horizon would double by 2028, rising to over three billion euros per year. This layer connects RAISE to the gigafactory network: computing capacity gives real substance to the network’s promise.
‘There are still no ready answers to questions of scientific integrity, credibility, and reproducibility’
The biggest innovation, however, is integrating public and private resources. Among the first companies to join the AI in Science Enablers initiative launched at the Dublin summit are Elsevier, which promises free AI tools and long-term research support; Mistral, which would provide computing access via a large-scale, sovereign infrastructure; as well as Nvidia and Google DeepMind. According to the Commission, CERN would work with the French company Mistral to build the largest foundational model in physics to date, which gives a good sense of the scale involved. These commitmentstherefore, involve not only funding but also tools, models, and capacity.
However, this is precisely where the initiative’s most sensitive issue lies. If Europe’s scientific artificial intelligence is based in part on American tools—such as those from Nvidia and Google—or on models owned by private companies, then the issue of sovereignty, discussed so often over the past year, resurfaces even in the laboratories. Another sensitive area is ethics: the Commission has updated its living guidelines on the responsible use of generative artificial intelligence in research, now extending them to include autonomous, agent-based systems, and is also seeking expert input on the scientific use of AI. Critics, however, argue that while the strategy is bold in rhetoric, it is more cautious on specific data and ethical mechanisms, and that no ready answers exist for questions of scientific integrity, credibility, and reproducibility.
Meanwhile, global competition is unfolding in the background. While the United States focuses on scale and self-regulation and China is mobilizing state resources, Europe would build its path on jointly built infrastructure, open science, and reliable frameworks. RAISE is the scientific pillar of this strategy: it does not seek to compete in building the largest models, but rather in empowering researchers to put existing tools at the service of discovery.
‘[RAISE] tests whether Europe can translate its scientific excellence into AI leadership without sacrificing its values or autonomy’
For researchers in Hungary and Central and Eastern Europe, this is not an abstract issue. Smaller member states connect to the continent’s AI capacity through so-called AI access points; in Hungary’s case, HunAIFAled by HUN-REN SZTAKI, provides a national access point to one of Europe’s largest supercomputers. The RAISE gigafactory access layer makes this gateway valuable: for emerging research systems, it offers a practical opportunity to avoid being left behind at the forefront of scientific AI. The stakes are twofold: access and ensuring the region participates, not merely uses.
RAISE is thus much more than just another EU acronym. It tests whether Europe can translate its scientific excellence into AI leadership without sacrificing its values or autonomy. The initiative is headed in the right direction; the question is implementation: will it be built quickly enough and with sufficient resources to turn the promise of a network into a large-scale reality before global competition is decided?
Related articles:
Click here to read the original article.
At a time when public debate is increasingly polarized and superficial, Hungarian Conservative remains committed to depth, intellectual honesty, and independent conservative thought.
Producing high-quality journalism requires resources. Your contribution helps us expand our coverage, reach new audiences, and keep our content accessible.
Source: www.hungarianconservative.com




