AI in Science Strategy EU Puts €600 Million Behind Access to AI Gigafactories

The European Union is putting up to €600 million from Horizon Europe toward dedicated access to AI Gigafactories for European researchers and startups, as part of its broader strategy to make high-performance computing more accessible for scientific research. The initiative sits within the EU’s Resource for AI Science in Europe (RAISE), which is designed to combine computing power, data, talent and research funding. With AI increasingly used in areas ranging from drug discovery and materials science to climate modelling, the EU is building infrastructure intended to give scientists access to much larger computing resources.

Europe Wants AI Computing to Become Scientific Infrastructure

Artificial intelligence is becoming increasingly embedded in scientific research.

Researchers are using machine-learning systems to analyse biological data, predict molecular structures, design materials, process astronomical observations and build increasingly detailed models of complex physical systems.

But sophisticated AI research requires something that is often in short supply: computing power.

Training and running advanced AI models can require large clusters of specialised processors, substantial data-storage capacity and high-speed networking. For individual universities, research groups and startups, gaining access to that infrastructure can be difficult and expensive.

The European Commission’s European AI in Science Strategy, presented in October 2025, is designed to address that challenge by creating a coordinated ecosystem for AI-enabled scientific research. At its centre is RAISE — the Resource for AI Science in Europe, a virtual institute intended to pool computational power, data, talent and research funding across Europe.

One of the strategy’s most significant infrastructure commitments is up to €600 million from Horizon Europe toward dedicated access to AI Gigafactories for EU researchers and startups.

What the €600 Million Is Intended to Do

The €600 million should not be understood simply as a conventional grant programme handing €600 million directly to individual scientists.

The European Commission describes it as an investment intended to secure dedicated access to AI Gigafactories for scientists and startups, including projects linked to Horizon Europe objectives.

That distinction matters because the infrastructure itself is becoming a strategic component of European science policy.

AI Gigafactories are being designed as very large-scale computing facilities capable of supporting the development and training of advanced AI systems. The European Commission describes them as the next step beyond today’s AI Factories, combining massive computing capacity with energy-efficient data centres, advanced networking and AI-driven automation.

For scientific researchers, access to this infrastructure could reduce one of the major barriers to experimenting with increasingly computationally intensive AI models.

RAISE Connects Computing, Data and Scientific Talent

The EU’s approach is broader than simply adding more processors.

RAISE is conceived as a virtual research institute that brings together several resources required for AI-enabled science: computational capacity, datasets, researchers, expertise and funding.

The pilot is already receiving Horizon Europe funding. The European Commission says the RAISE pilot has dedicated funding of €33 million under the 2025 work programme and €107 million under the 2026–27 work programme.

The initiative covers scientific fields including fundamental physics and astronomy, materials science, life sciences, Earth sciences, and social sciences and humanities.

That cross-disciplinary structure is important because advances in AI research can often be transferred between fields.

A technique developed for analysing biological sequences, for example, could inspire approaches to other forms of complex scientific data. Likewise, improvements in AI model efficiency can benefit researchers regardless of whether they are studying molecules, climate systems or astronomical observations.

Why AI Needs High-Performance Computing

Modern AI models can involve enormous numbers of parameters and require significant computational resources for training and inference.

Scientific applications can be even more demanding because researchers may need to combine AI models with large datasets, simulations and specialised computational tools.

Consider drug discovery.

An AI system may need to analyse molecular structures, predict interactions and evaluate huge numbers of possible candidates. Materials researchers may use AI to explore combinations of chemical compositions and physical properties. Climate scientists can combine machine learning with large-scale environmental datasets and numerical models.

The availability of more computing capacity does not automatically solve these problems. Researchers still need suitable algorithms, high-quality data and scientific validation.

But without sufficient computing resources, many potentially useful approaches cannot be tested at meaningful scale.

The European Commission therefore identifies improved access to computational power as one of the key enablers of its AI-in-science strategy.

AI Factories Are Already Building the Foundation

The planned Gigafactories are not emerging in isolation.

Europe has already been developing a network of AI Factories around its EuroHPC supercomputing infrastructure.

According to the European Commission, 19 AI Factories and 13 AI Factory Antennas are being established across Europe. The network is intended to provide computing and support services to researchers, startups, SMEs, industry and public authorities.

These facilities are connected to Europe’s broader supercomputing infrastructure.

The idea is to bring together computing power, data and technical expertise so that organisations that do not possess massive computing infrastructure themselves can still develop and test AI systems.

The Gigafactories represent a much larger scale.

The Commission says the EU has launched a call to establish up to seven AI Gigafactories across Europe. These facilities are expected to contain more than 100,000 advanced AI processors each and support the development and training of next-generation AI models containing trillions of parameters.

A Larger Infrastructure Push Is Underway

The AI Gigafactory initiative is also part of a much larger European investment programme.

In July 2026, the European Commission announced the launch of the call for AI Gigafactories, saying the initiative could involve up to €10 billion in EU and national funding and was expected to unlock at least €20 billion in private investment across the European Union.

The infrastructure is intended to serve more than AI research alone.

Startups, scaleups, SMEs, industrial companies, universities, researchers and public authorities are all expected to gain access to the computing infrastructure for activities including AI model training, inference and fine-tuning.

This creates a potentially important connection between scientific research and Europe’s emerging AI industry.

A university research group could use the same broader infrastructure ecosystem as a startup developing a scientific AI product, while industrial researchers could potentially access the computing capacity required for large-scale experimentation.

Scientific Research Could Become a Major AI Use Case

The European Commission’s strategy is based on the idea that AI is not merely another digital productivity tool.

It can change how scientific research itself is performed.

AI can assist with literature analysis, laboratory automation, biological modelling, environmental prediction and other stages of the scientific process. The Commission specifically points to applications ranging from improving cancer treatments to environmental research and earthquake-impact prediction.

The goal is therefore not simply to develop larger AI models.

It is to create an infrastructure in which scientists can use AI as part of the research process.

That includes both AI for science — using AI to accelerate scientific discovery — and science for AI, where scientific research helps advance the capabilities of AI systems themselves.

RAISE is explicitly designed around both directions.

From Drug Discovery to Climate Science

The potential applications cover a wide range of scientific disciplines.

Life Sciences

AI can analyse genomic and molecular datasets, model biological structures and help identify potential therapeutic targets. Larger computing resources can allow researchers to test more sophisticated models and process larger datasets.

Materials Science

Researchers can use AI to search through enormous combinations of materials and predict properties before conducting physical experiments. This could help narrow down candidates for batteries, catalysts, electronic materials and other technologies.

Earth and Climate Science

Machine learning can process satellite observations, environmental measurements and simulation outputs. AI can also help researchers identify patterns across datasets that are difficult to analyse using conventional approaches alone.

Physics and Astronomy

Large scientific datasets are increasingly generated by observatories, particle experiments and other instruments. AI can help classify events, detect patterns and accelerate parts of computational analysis.

The European Commission’s RAISE framework includes these fields among its initial scientific areas of focus.

The Startup Dimension Is Equally Important

The strategy also has implications for smaller companies.

Large AI infrastructure is expensive, which can create a structural advantage for companies and institutions that already have access to major computing clusters.

By providing access through shared European infrastructure, the EU is attempting to reduce that barrier for startups and smaller research organisations.

The European Commission explicitly identifies startups and SMEs as users of the wider AI Factory ecosystem. The Gigafactory programme is similarly designed to provide infrastructure access to startups, scaleups and SMEs alongside academia and industry.

For scientific startups, access to large-scale computing can be particularly important.

A company developing an AI system for molecular discovery, advanced materials or scientific simulation may need substantial computational resources before it can demonstrate that its technology works at scale.

Shared infrastructure can potentially allow those companies to experiment without first building an entire supercomputing environment themselves.

More Computing Does Not Mean Automatic Scientific Breakthroughs

The EU’s investment does not guarantee that AI will produce faster scientific discoveries in every field.

Computing power is only one part of the research equation.

AI systems still depend on reliable datasets, appropriate scientific methods and rigorous validation. A model can identify a correlation without establishing a physical mechanism. A prediction can look convincing while failing under laboratory testing.

The European Commission itself emphasises responsible AI use in science, including scientific integrity and methodological rigour. Its AI-in-science strategy aims not only to increase adoption but also to address science-specific challenges associated with AI.

That means the infrastructure expansion is being paired with efforts around talent, data and responsible research practices.

Europe Is Building a Scientific AI Stack

The emerging European model can be viewed as a stack of interconnected resources.

At the bottom is high-performance computing.

Above that sit AI Factories and the planned Gigafactories, providing access to specialised computing infrastructure. RAISE adds coordination between researchers, datasets, funding and expertise. Scientific laboratories and universities provide domain knowledge, while startups and industry can turn research results into applications.

The strategy also connects this infrastructure with Europe’s wider goal of strengthening technological capacity within the region.

The European Commission says the AI Gigafactory initiative is intended to strengthen Europe’s technological leadership, resilience and strategic autonomy by providing advanced AI infrastructure on European soil.

For science, the significance is potentially substantial.

If researchers can obtain reliable access to much larger computing resources without individually building those systems, the cost and time required to experiment with AI-based scientific methods could fall.

A New Phase for AI-Powered Science

Europe’s €600 million Horizon Europe commitment for dedicated AI Gigafactory access is part of a much broader transformation in how scientific computing is being organised.

The objective is not simply to build larger AI systems.

It is to make advanced computing available as a research resource — connecting scientists and startups with the processors, data, expertise and funding needed to develop AI-enabled scientific applications.

The scale of the planned infrastructure reflects how quickly AI is becoming part of scientific research.

As AI models become more capable, the limiting factor may increasingly shift from whether researchers can imagine a useful application to whether they have enough computing power, data and specialised expertise to test it.

Europe’s response is to build that infrastructure before the demand reaches its full scale.

If the strategy succeeds, AI Gigafactories could become more than massive computing centres. They could form part of a shared scientific infrastructure through which European researchers and companies develop new approaches to biology, materials, climate science, physics and other fields.

The scientific results, however, will ultimately depend on what researchers are able to build with that infrastructure — and how effectively AI can be integrated with the experimental methods that turn computational predictions into verified discoveries.

Most Searched 5 FAQs

1. How much is the EU investing in AI for science?
The European Commission says Horizon Europe will invest up to €600 million to secure dedicated access to AI Gigafactories for EU scientists and startups as part of the European AI in Science Strategy.

2. What are AI Gigafactories?
AI Gigafactories are planned large-scale computing facilities designed to develop, train and deploy advanced AI models. The EU describes them as infrastructure combining very large numbers of AI processors with high-speed networking, energy-efficient data centres and AI-driven automation.

3. What is RAISE in the European AI in Science Strategy?
RAISE, or the Resource for AI Science in Europe, is a virtual institute designed to pool computational power, data, talent and research funding to support AI-enabled scientific research across Europe.

4. Who will be able to use Europe’s AI Gigafactories?
The wider AI Gigafactory initiative is intended to provide infrastructure for startups, scaleups, SMEs, industry, academia, researchers and public authorities. The AI-in-science strategy specifically aims to secure dedicated access for EU researchers and startups.

5. What scientific fields could benefit from AI Gigafactories?
Potential areas include life sciences, materials science, physics, astronomy, Earth sciences and environmental research. The RAISE programme is being developed across several of these fields while providing computing, data and research resources for AI-enabled science.