Live

As artificial intelligence continues to reshape how research is done and discoveries are made, the European Union is stepping up to define what a uniquely European path for AI in science should look like. At the second AI4Science workshop, held on 12 June 2025 in Seville, the European Commission brought researchers, policymakers, and innovators together to co-create this vision.

Organised by the Joint Research Centre (JRC) and the Directorate-General for Research and Innovation (DG RTD), the event went beyond exchanging opinions—it marked a serious step toward developing a policy framework that combines scientific excellence with ethical responsibility.

Following recent online public consultations, which gathered nearly 600 responses and 166 written opinions from across Europe, the workshop offered an in-depth, human-centred look at what needs to happen next. The results will feed into a strategic document to be published by the Commission before the end of 2025.

A double goal: AI in science and AI for science

At the heart of this strategy is the idea of AI Science, explained by Lars de Nul from DG RTD. It’s about two things: using AI as a powerful tool for research, and driving the development of AI itself through science.

This means Europe isn’t just trying to compete with the US or China on speed and scale. Instead, it’s aiming to become a leader in responsible innovation — where AI works for society, supports researchers, and never undermines public trust in science.

Participants at the workshop emphasized that Europe’s true strength lies in its ability to create a smart, human-first AI ecosystem grounded in shared values — not just in developing the fastest tools.

Don’t let talent slip away

One of the hottest topics? People. Many experts warned that Europe risks losing brilliant minds to countries with more attractive academic offers — especially the US and China. To reverse that trend, Europe needs:

  • Stronger research institutions,
  • Interdisciplinary career paths,
  • High-quality training programmes.

Another concern was the messy flood of unverified, poor-quality educational materials about AI. Clear standards and trustworthy references are essential to help young researchers find reliable knowledge and avoid misinformation.

Data infrastructure: a roadblock or a launchpad?

“No AI without data.” That phrase was repeated throughout the Seville workshop. Participants strongly supported boosting funding for datasets built specifically for scientific use. They also called for better metadata standards, system compatibility, and tools to track data origins.

But integrating all that data into existing infrastructures — like Europe’s “AI Factories” or the EuroHPC platform — isn’t easy. Technical barriers and fragmented systems mean that Europe still isn’t getting the most out of its scientific resources.

The solution? More coordination and a system where computing power and high-quality data are shared more openly and efficiently.

Trust, ethics, and the risk of “bad AI”

Ethics and trust kept coming up. Without strong safeguards, experts warned, AI tools — especially generative ones — could damage the credibility of science. The risks include fake content, plagiarism, and the misuse of research.

Clear EU-wide ethical guidelines are essential, not just as nice declarations but as real-world rules applied in everyday research. Transparency, explainability, privacy, and sustainability were named as the four cornerstones of trustworthy AI in science.

Diversity is good, coordination is better

Europe’s national AI strategies and diverse research communities are a strength — but only if they work together. Many voices in Seville argued that the EU needs more coherence: fewer duplicated efforts, more cross-border collaboration, and aligned national policies.

Institutions like the JRC could play a key role in coordinating a pan-European approach to AI in science. The bottom line? If Europe wants to lead in trusted AI, it must speak with one voice — respecting diversity, but united by shared goals.

Leading by example, not just competing

The AI4Science workshop made it clear: Europe doesn’t just want to play catch-up in the global AI race. It wants to lead — on its own terms.

Not by chasing patents or raw computing power, but by setting global standards for ethical, human-centric, and socially responsible AI.

In a world where trust in science and tech is becoming harder to maintain, this might just be Europe’s most powerful asset.

The ideas gathered in Seville will directly shape the EU’s future strategy on AI in science. And that strategy will matter not only for Europe’s place in global research rankings, but for the kind of future young people across the continent will help shape.

Written by

Shape the conversation

Do you have anything to add to this story? Any ideas for interviews or angles we should explore? Let us know if you’d like to write a follow-up, a counterpoint, or share a similar story.

Write a counterargument article Write a related article